Balogun1996Interventions For Promoting The Initiation of Breastfeeding

Cochrane Database of Systematic Reviews
Interventions for promoting the initiation of breastfeeding

(Review)
Balogun OO, O’Sullivan EJ, McFadden A, Ota E, Gavine A, Garner CD, Renfrew MJ, MacGillivray S
Balogun OO, O’Sullivan EJ, McFadden A, Ota E, Gavine A, Garner CD, Renfrew MJ, MacGillivray S.
Interventions for promoting the initiation of breastfeeding.
Cochrane Database of Systematic Reviews 2016, Issue 11. Art. No.: CD001688.
DOI: 10.1002/14651858.CD001688.pub3.
www.cochranelibrary.com
Interventions for promoting the initiation of breastfeeding (Review)

Copyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
TABLE OF CONTENTS
HEADER . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
ABSTRACT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
PLAIN LANGUAGE SUMMARY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
SUMMARY OF FINDINGS FOR THE MAIN COMPARISON . . . . . . . . . . . . . . . . . . . 4
BACKGROUND . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
OBJECTIVES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
METHODS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
RESULTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
Figure 1. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
Figure 2. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
Figure 3. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
ADDITIONAL SUMMARY OF FINDINGS . . . . . . . . . . . . . . . . . . . . . . . . . . 16
DISCUSSION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
AUTHORS’ CONCLUSIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
ACKNOWLEDGEMENTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
REFERENCES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
CHARACTERISTICS OF STUDIES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
DATA AND ANALYSES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89
Analysis 1.1. Comparison 1 Healthcare professional-led breastfeeding education and support versus standard care, Outcome
1 Initiation of breastfeeding. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91
Analysis 2.1. Comparison 2 Non-healthcare professional-led breastfeeding education and support versus standard care,
Outcome 1 Initiation of breastfeeding. . . . . . . . . . . . . . . . . . . . . . . . . . 92
Analysis 2.2. Comparison 2 Non-healthcare professional-led breastfeeding education and support versus standard care,
Outcome 2 Early initiation of breastfeeding. . . . . . . . . . . . . . . . . . . . . . . . 93
Analysis 3.1. Comparison 3 Healthcare professional-led breastfeeding education with non-healthcare professional support
versus standard care, Outcome 1 Initiation of breastfeeding. . . . . . . . . . . . . . . . . . . 94
Analysis 4.1. Comparison 4 Healthcare professional-led breastfeeding education with peer support versus attention control,
Outcome 1 Initiation of breastfeeding. . . . . . . . . . . . . . . . . . . . . . . . . . 94
Analysis 5.1. Comparison 5 Breastfeeding education using multimedia versus routine care, Outcome 1 Initiation of
breastfeeding. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95
Analysis 6.1. Comparison 6 Early mother-infant contact versus standard care, Outcome 1 Initiation of breastfeeding. 96
Analysis 7.1. Comparison 7 Community-based breastfeeding groups versus no breastfeeding groups, Outcome 1
Breastfeeding rate at birth. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96
APPENDICES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97
WHAT’S NEW . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97
HISTORY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97
CONTRIBUTIONS OF AUTHORS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98
DECLARATIONS OF INTEREST . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99
SOURCES OF SUPPORT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99
DIFFERENCES BETWEEN PROTOCOL AND REVIEW . . . . . . . . . . . . . . . . . . . . . 99
INDEX TERMS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99
Interventions for promoting the initiation of breastfeeding (Review) i

[Intervention Review]
Interventions for promoting the initiation of breastfeeding
Olukunmi O Balogun1 , Elizabeth J O’Sullivan2 , Alison McFadden3 , Erika Ota4 , Anna Gavine5 , Christine D Garner6 , Mary J Renfrew
3
, Stephen MacGillivray5
1 Department of Health Policy, National Center for Child Health and Development, Tokyo, Japan. 2 School of Medicine, University
College Dublin, Dublin, Ireland. 3 Mother and Infant Research Unit, School of Nursing and Health Sciences, University of Dundee,
Dundee, UK. 4 Global Health Nursing, St. Luke’s International University, Graduate School of Nursing Sciences, Tokyo, Japan.
5 evidence Synthesis Training and Research Group (eSTAR), University of Dundee, Dundee, UK. 6 Division of Nutritional Sciences,
Cornell University, Ithaca, NY, USA
Contact address: Olukunmi O Balogun, Department of Health Policy, National Center for Child Health and Development, 2-10-1
Okura, Setagaya, Tokyo, Tokyo, 157-8535, Japan. omobolanle_balogun@yahoo.com, balogun-o@ncchd.go.jp.
Editorial group: Cochrane Pregnancy and Childbirth Group.

Publication status and date: New search for studies and content updated (no change to conclusions), published in Issue 11, 2016.
Review content assessed as up-to-date: 29 February 2016.
Citation: Balogun OO, O’Sullivan EJ, McFadden A, Ota E, Gavine A, Garner CD, Renfrew MJ, MacGillivray S. Interventions
for promoting the initiation of breastfeeding. Cochrane Database of Systematic Reviews 2016, Issue 11. Art. No.: CD001688. DOI:
10.1002/14651858.CD001688.pub3.
ABSTRACT
Background
Despite the widely documented risks of not breastfeeding, initiation rates remain relatively low in many high-income countries,
particularly among women in lower-income groups. In low- and middle-income countries, many women do not follow World Health
Organization (WHO) recommendations to initiate breastfeeding within the first hour after birth. This is an update of a Cochrane
Review, first published in 2005.
Objectives
To identify and describe health promotion activities intended to increase the initiation rate of breastfeeding.
To evaluate the effectiveness of different types of breastfeeding promotion activities, in terms of changing the number of women who
initiate breastfeeding.
To evaluate the effectiveness of different types of breastfeeding promotion activities, in terms of changing the number of women who
initiate breastfeeding early (within one hour after birth).
Search methods
We searched Cochrane Pregnancy and Childbirth’s Trials Register (29 February 2016) and scanned reference lists of all articles obtained.
Selection criteria
Randomised controlled trials (RCTs), with or without blinding, of any breastfeeding promotion intervention in any population group,
except women and infants with a specific health problem.
Data collection and analysis
Two review authors independently assessed trial reports for inclusion, extracted data and assessed trial quality. Discrepancies were
resolved through discussion and a third review author was involved when necessary. We contacted investigators to obtain missing
information.
Interventions for promoting the initiation of breastfeeding (Review) 1
Main results
Twenty-eight trials involving 107,362 women in seven countries are included in this updated review. Five studies involving 3,124
women did not contribute outcome data and we excluded them from the analyses. The methodological quality of the included trials
was mixed, with significant numbers of studies at high or unclear risk of bias due to: inadequate allocation concealment (N = 20); lack
of blinding of outcome assessment (N = 20); incomplete outcome data (N = 19); selective reporting (N = 22) and bias from other
potential sources (N = 17).
Healthcare professional-led breastfeeding education and support versus standard care
The studies pooled here compare professional health workers delivering breastfeeding education and support during the prenatal and
postpartum periods with standard care. Interventions included promotion campaigns and counselling, and all took place in a formal
setting. There was evidence from five trials involving 564 women for improved rates ofbreastfeeding initiation among women who
received healthcare professional-led breastfeeding education and support (average risk ratio (RR) 1.43, 95% confidence interval (CI)
1.07 to 1.92; Tau² = 0.07, I² = 62%, low-quality evidence) compared to those women who received standard care. We downgraded
evidence due to design limitations and heterogeneity. The outcome of early initiation of breastfeeding was not reported in the studies
under this comparison.
Non-healthcare professional-led breastfeeding education and support versus standard care
There was evidence from eight trials of 5712 women for improved rates of breastfeeding initiation among women who received
interventions from non-healthcare professional counsellors and support groups (average RR 1.22, 95% CI 1.06 to 1.40; Tau² = 0.02,
I² = 86%, low-quality evidence) compared to women who received standard care. In three trials of 76,373 women, there was no clear
difference between groups in terms of the number of women practicing early initiation of breastfeeding (average RR 1.70, 95% CI
0.98 to 2.95; Tau² = 0.18, I² = 78%, very low-quality evidence). We downgraded the evidence for a combination of design limitations,
heterogeneity and imprecision (wide confidence intervals crossing the line of no effect).
Other comparisons
Other comparisons in this review also looked at the rates of initiation of breastfeeding and there were no clear differences between
groups for the following comparisons of combined healthcare professional-led education with peer support or community educator
versus standard care (2 studies, 1371 women) or attention control (1 study, 237 women), breastfeeding education using multimedia
(a self-help manual or a video) versus routine care (2 studies, 497 women); early mother-infant contact versus standard care (2 studies,
309 women); and community-based breastfeeding groups versus no breastfeeding groups (1 study, 18,603 women). None of these
comparisons reported data on early initiation of breastfeeding.
Authors’ conclusions
This review found low-quality evidence that healthcare professional-led breastfeeding education and non-healthcare professional-led
counselling and peer support interventions can result in some improvements in the number of women beginning to breastfeed. The
majority of the trials were conducted in the USA, among women on low incomes and who varied in ethnicity and feeding intention,
thus limiting the generalisability of these results to other settings.
Future studies would ideally be conducted in a range of low- and high-income settings, with data on breastfeeding rates over various
timeframes, and explore the effectiveness of interventions that are initiated prior to conception or during pregnancy. These might
include well-described interventions, including health education, early and continuing mother-infant contact, and initiatives to help
mothers overcome societal barriers to breastfeeding, all with clearly defined outcome measures.
PLAIN LANGUAGE SUMMARY

Interventions for encouraging women to start breastfeeding
What is the issue?
International rates of breastfeeding initiation are extremely variable both between and within countries. Low- and middle-income
countries generally have high rates of women starting breastfeeding, and the challenge is for breastfeeding to begin within one hour
of birth. High-income countries have a much greater variation in the number of women who start breastfeeding, with more highly
educated and more well-off women likely to start.
The World Health Organization recommends that breastfeeding should start within the first hour after giving birth, that all infants
should be exclusively breastfed from birth to six months of age, and that breastfeeding should continue until 2 years or beyond. We
know that breastfeeding is good for the health of women and babies. Babies who are not fully breastfed for the first three to four months
of life are more likely to suffer from infections of the stomach and intestines, air passages and lungs, or develop ear infections. Babies
who are not breastfed are more likely to be overweight or have diabetes later in life, and mothers who do not breastfeed have increased
risks of breast and ovarian cancer. Other practical benefits of breastfeeding include saving money on buying breast milk substitutes and,
for society, on treating illness. Yet many women feed their babies with infant formula.
Why is this important?
We want to have a better understanding of what works to promote breastfeeding, for women, their families, the health system and
society. Women face many barriers to breastfeeding, including lack of public spaces where women can breastfeed without feeling
embarrassment; lack of flexible working days for breastfeeding women at work; widespread advertising of breast milk substitutes; and
public policy that ignores the needs of breastfeeding women. New ways to promote breastfeeding are needed.
What evidence did we find?
We searched for evidence on 29 February 2016. This updated review now includes 28 randomised controlled studies involving 107,362
women. Twenty studies involving 27,865 women looked at interventions to increase the number of women who started breastfeeding, in
three high-income countries (Australia, 1 study; UK, 4 studies; and USA, 14 studies) and one lower middle-income country (Nicaragua,
1 study). Three studies investigated the effect of an intervention to increase the number of women who started breastfeeding early,
within one hour after birth. These involved 76,373 women from Malawi, Nigeria and Ghana. The study from Malawi was large, with
55,931 participants.
Health education delivered by doctors and nurses and counselling and peer support by trained volunteers improved the number of
women who began breastfeeding their babies. Five studies involving 564 women reported that women who received breastfeeding
education and support from doctors or nurses were more likely to start breastfeeding compared to women who received standard care.
Four of these studies were conducted in low-income or amongst minority ethnic women in the USA, where baseline breastfeeding
rates are typically low. Eight studies involving 5712 women showed improved rates of starting breastfeeding with trained volunteer-
delivered interventions and support groups compared to the women who received standard care.
Breastfeeding education provided by trained volunteers could also improve the rates of early initiation of breastfeeding, within one
hour of giving birth, in low-income countries.
We assessed all the evidence in this review to be low-quality because of limitations in study design and variations in the interventions, to
whom, when, where, and how an intervention was delivered. Standard care also differed and could include some breastfeeding support,
for example, in the UK.
We found too little evidence to say whether strategies with multimedia, early mother-infant contact, or community-based breastfeeding
groups were able to improve breastfeeding initiation.
What does this mean?
Doctors, nurses, and trained volunteers can deliver education sessions and provide counselling and peer support to increase the number
of women who start breastfeeding their babies. High-quality research is needed to understand which interventions are likely to be
effective in different population groups. More studies are needed in low- and middle-income countries to find out which strategies will
encourage women to start breastfeeding just after giving birth.

Interventions for promoting the initiation of breastfeeding (Review) S U M M A R Y O F F I N D I N G S F O R T H E M A I N C O M P A R I S O N [Explanation]
Population: wom en exposed to interventions intended to prom ote breastf eeding

Setting: USA, Ireland
Intervention: healthcare prof essional-led breastf eeding education and support
Comparison: standard care
Outcomes Anticipated absolute effects∗ (95% CI) Relative effect of participants Quality of the evidence Comments
(95% CI) (studies) (GRADE)
Risk with standard care Risk with health-

care professional- led
breastfeeding educa-
tion and support
Initiation of breastf eed- Study population average RR 1.43 564 ⊕⊕ It is not possible to
ing (1.07 to 1.92) (5 RCTs) LOW 1,2 blind this type of in-
tervention and so we
418 per 1000 598 per 1000 have not downgraded
(448 to 808) f or lack of blinding
Early initiation of No trial included in this com parison m easured the outcom e of early initiation of breastf eeding
breastf eeding
* The risk in the intervention group (and its 95% conf idence interval) is based on the assum ed risk in the com parison group and the relative effect of the intervention (and its
95% CI).
CI: conf idence interval; RCT: random ised controlled trial; RR: risk ratio.
GRADE Working Group grades of evidence

High quality: We are very conf ident that the true ef f ect lies close to that of the estim ate of the ef f ect.
M oderate quality: We are m oderately conf ident in the ef f ect estim ate: The true ef f ect is likely to be close to the estim ate of the ef f ect, but there is a possibility that it is
substantially dif f erent.
Low quality: Our conf idence in the ef f ect estim ate is lim ited: The true ef f ect m ay be substantially dif f erent f rom the estim ate of the ef f ect.
Very low quality: We have very little conf idence in the ef f ect estim ate: The true ef f ect is likely to be substantially dif f erent f rom the estim ate of ef f ect
1
M ost studies were unclear f or allocation concealm ent and som e studies were of high risk f or attrition bias. Downgraded f or
risk of bias (-1).
2 High heterogeneity (I² > 60%) (-1).
4
BACKGROUND (Victora 2016b). However, country-level breastfeeding rates con-
ceal geographical and social gradients in breastfeeding initiation.
For example, in the UK, breastfeeding initiation rates range from
Description of the condition 83% in England to 64% in Northern Ireland (McAndrew 2012).
At the same time, there is a stark social-class gradient with the high-
There is extensive, good-quality evidence for short-term and long-
est incidence of breastfeeding in women aged over 30 years (87%),
term health risks of formula-feeding. The World Health Orga-
those who continued education beyond 18 years (91%), and those
nization (WHO) recommends initiation of breastfeeding within
in managerial and professional occupations (90%) (McAndrew
the first hour after birth, exclusive breastfeeding for the first six
2012). In the USA, the lowest rates of breastfeeding initiation are
months, with continued breastfeeding along with appropriate
among black women (54%) (CDCP 2010).
complementary foods up to two years of age or beyond (WHO
One of the major factors contributing to low rates of breastfeed-
2003). Babies who are not breastfed are more likely to suffer in-
ing initiation is the influence of the breast milk substitute indus-
fectious diseases such as gastroenteritis, respiratory disease, and
try. It has been estimated that the retail value of the industry will
otitis media (middle-ear infections) leading to increased hospi-
reach USD 70.6 billion by 2019 (Rollins 2016). Inadequate im-
talisation, morbidity, and mortality (Bowatte 2015; Horta 2013;
plementation and enforcement of The International Code of Mar-
Sankar 2015). Children who have not been breastfed have in-
keting of Breast Milk Substitutes (WHO 1981) is one key factor
creased rates of childhood diabetes and obesity (Horta 2015a), and
influencing women’s decision to breastfeed, and the belief that, in
increased dental disease (Peres 2015; Tham 2015). In addition,
spite of the evidence to the contrary, infant formula has equivalent
there is evidence of an adverse impact of not being breastfed on IQ,
nutritional value to breast milk (McFadden 2016). It is unclear
and educational and behavioural outcomes for the child (Heikkilä
whether the availability of subsidised infant formula milk through
2011; Heikkilä 2014; Horta 2015b; Quigley 2012). For women,
welfare food programmes, such as the UK-based Healthy Start
good-quality evidence shows associations between not breastfeed-
Programme and the USA-based Special Supplemental Nutrition
ing and increased risks of breast and ovarian cancer, and diabetes
Program for Women, Infants, and Children, is an economic fac-
(Chowdhury 2015). For preterm babies, a diet of exclusive breast
tor which contributes unintentionally to women in low-income
milk reduces the incidence of necrotising enterocolitis - a disease
groups deciding to formula feed (see for example Jiang 2010).
of the gastrointestinal tract of premature infants that results in
inflammation and bacterial invasion of the bowel wall (Hermann
2014; Ip 2007).
Attempts have been made to quantify public cost benefits of breast-
Description of the intervention
feeding. The global cost burden of not breastfeeding was estimated The decision to breastfeed is influenced by multiple complex fac-
by Rollins 2016 to be USD 302 billion annually. In the UK, tors at the individual, family, health system, and societal levels
Renfrew 2012a estimated that a modest increase in breastfeeding (Dyson 2010). Consequently, there are many approaches to pro-
rates could save over GBP 17 million per annum by avoiding the moting the initiation of breastfeeding which may target preg-
costs of treating four acute diseases in infants (gastrointestinal in- nant women, their families, wider communities and society, or the
fection, lower respiratory tract infection, otitis media, and necro- health service. Interventions to promote the initiation of breast-
tising enterocolitis). feeding are delivered before the first feed, i.e. before or during
International rates of initiation of breastfeeding are extremely vari- pregnancy, or immediately after birth.
able between and within countries. As data are gathered using Interventions targeted to individual women include health edu-
different methods in different settings, reported rates should be cation, peer support, practical skills training and early mother-
treated with caution. From countries where data are available, low- and-baby contact. Health education interventions to promote the
and middle-income countries generally have high rates of breast- initiation of breastfeeding delivered during pregnancy may entail
feeding initiation of over 90% (Victora 2016b). However there one or more sessions, be delivered to groups or one-to-one, in for-
is often a delay in initiating breastfeeding beyond the first hour mal or informal settings, and be delivered by health professionals,
after birth, which increases neonatal mortality (NEOVITA Study maternity support workers, or peer supporters who may be trained
Group 2016). The world average for early initiation of breast- or untrained. Breastfeeding health education may be targeted to
feeding is 44% (UNICEF 2014), however there is wide variation, women alone or it may include family members such as partners
with some countries such as India and Pakistan reporting rates of and parents (Grassley 2007; Ingram 2004). The content of health
23.3% and 18.4% respectively (Victora 2016b). In high-income education to promote the initiation of breastfeeding may include
countries, there is wide variation of breastfeeding initiation. Many the health outcomes of breastfeeding compared to formula-feed-
countries report rates of over 90% such as Australia, Chile, the ing, what to expect when breastfeeding, and how to prevent and
Nordic countries, Italy, Japan, Russia, and Saudi Arabia (Victora solve breastfeeding-related problems. It may also include practi-
2016b); however, lower rates are reported from the UK (81%), the cal skills such as positioning and attachment of the baby at the
USA (79%), France (63%), and the Republic of Ireland (55%) breast, and the opportunity to talk to a breastfeeding woman and

observe a breastfeed. There is increasing focus on health educa- formula-feeding culture. Structured programmes such as BFHI/
tion approaches to predict and support behaviour change, such as BFI work through addressing many of the negative influences on
motivational interviewing and the Theory of Planned Behaviour women’s infant-feeding decisions that derive from health service
(see for example Copeland 2015; Lawton 2012). policy and the knowledge, skills, and attitudes of health personnel
Peer support interventions to promote the initiation of breast- (Rollins 2016). Not least of these is protecting women and staff
feeding are generally targeted at communities where breastfeeding from the influence of marketing and promotion of breast milk
rates are low, and involve contact between a pregnant woman and substitutes (Piwoz 2015).
a woman from a similar background who has experience of breast-
feeding (Phipps 2006). This type of mother-to-mother support
has been shown to increase breastfeeding initiation rates (Dyson
Why it is important to do this review
2006). Peer supporters undergo varying lengths and styles of train-
ing, can be paid or unpaid, and they can be integrated into the The purpose of this review is to examine interventions which aim
healthcare team or separate. to encourage women to breastfeed, to evaluate their effectiveness
The most effective health service intervention to promote the in terms of changes in the number of women who initiate breast-
initiation of breastfeeding is the WHO/United Nations Inter- feeding, and in terms of changing the number of women who
national Children’s Emergency Fund (UNICEF) Baby Friendly initiate breastfeeding early (within one hour after birth). It is im-
Hospital Initiative (BFHI), also known in some countries as the portant to do this review to inform the design of interventions
Baby Friendly Initiative (BFI). The BFHI/BFI is a multifaceted, to promote the initiation of breastfeeding. Increasing rates of ini-
structured programme that involves organisational change (Beake tiation of breastfeeding is the first step towards meeting WHO
2012). The BFHI/BFI comprises implementation of the Ten Steps recommendations for breastfeeding and realising the potential of
to Successful Breastfeeding (WHO/UNICEF 1989), that cover breastfeeding in improving health, reducing the economic burden
policy, staff training, promotion and support of breastfeeding, lim- of ill health, and reducing health inequalities. It is also important
iting use of infant formula, teats and pacifiers, and keeping moth- to undertake this review to find effective interventions to counter
ers and babies together (rooming-in) (Pérez-Escamilla 2016). Im- the promotion of breast milk substitutes by the infant formula in-
plementation of BFHI/BFI has increased breastfeeding initiation dustry. The amount of money invested by formula manufacturers
rates in Israel, Taiwan, UK, and USA (Beake 2012; Pérez-Escamilla is many times greater than the amount spent by governments on
2016). promoting breastfeeding (Lutter 2013). The published Cochrane
Mass media campaigns are interventions that are targeted toward Review on support for healthy breastfeeding mothers with healthy
wider society, and, when implemented alongside other interven- term babies found that interventions had more effect on increas-
tions have had some success at increasing breastfeeding initiation ing exclusive breastfeeding before four to six weeks and before
rates (Fairbank 2000). six months in settings where there were high background rates of
breastfeeding initiation compared to areas where there were low
or intermediate rates (Renfrew 2012b).
How the intervention might work
Interventions to promote the initiation of breastfeeding work in
different ways that are likely to be context-specific, to vary ac-
cording to individual needs and circumstances (Rollins 2016), OBJECTIVES
and to vary by each country’s economic status and breastfeeding
rates. Successful interventions work through addressing the many 1. To identify and describe health promotion activities
structural, societal, economic, and individual influences on the intended to increase the initiation rate of breastfeeding.
decision to breastfeed (Rollins 2016). These include increasing
women’s motivation to breastfeed, whether that be via providing 2. To evaluate the effectiveness of different types of
information about the health outcomes of breastfeeding, provid- breastfeeding promotion activities, in terms of changing the
ing women with the skills and confidence to commence breast- number of women who initiate breastfeeding.
feeding, or using more structured approaches such as motivational
3. To evaluate the effectiveness of different types of
interviewing that seek to ‘increase an individual’s belief that they
breastfeeding promotion activities, in terms of changing the
can achieve a desired outcome’ (Copeland 2015). Interventions
number of women who initiate breastfeeding early (within one
that focus on women’s families and wider communities attempt
hour after birth).
to change societal perceptions and norms regarding infant-feed-
ing (Rollins 2016), reducing the impact of these barriers. These
types of interventions are particularly important in communities
where breastfeeding rates are low and there is an entrenched infant METHODS

Criteria for considering studies for this review Electronic searches
We searched Cochrane Pregnancy and Childbirth’s Trials Register
by contacting their Information Specialist (29 February 2016).
Types of studies The Register is a database containing over 22,000 reports of con-
trolled trials in the field of pregnancy and childbirth. For full search
We included individual randomised controlled trials (RCTs) or
methods used to populate Pregnancy and Childbirth’s Trials Regis-
cluster-RCTs, with or without blinding. There was no limitation
ter, including the detailed search strategies for CENTRAL, MED-
of study by country of origin or language. We excluded quasi-
LINE, Embase, and CINAHL, the list of handsearched journals
randomised trials and cross-over trials. We also excluded abstracts
and conference proceedings, and the list of journals reviewed via
for which we could not find the full reports.
the current awareness service, please follow this link to the edi-
torial information about Cochrane Pregnancy and Childbirth in
the Cochrane Library and select the ’Specialized Register’ section
Types of participants from the options on the left side of the screen.
Women exposed to interventions intended to promote breastfeed- Briefly, Cochrane Pregnancy and Childbirth’s Trials Register is
ing. This includes pregnant women, mothers of newborn infants, maintained by their Information Specialist and contains trials
and women who may decide to breastfeed in the future. We also identified from:
included population subgroups of women, such as women from 1. monthly searches of the Cochrane Central Register of
low-income or ethnic groups. Women and infants with a specific Controlled Trials (CENTRAL);
health problem, e.g. mothers with HIV/AIDS or infants with cleft 2. weekly searches of MEDLINE (Ovid);
palate, or premature babies, are excluded from this review. 3. weekly searches of Embase (Ovid);
4. monthly searches of CINAHL (EBSCO);
5. handsearches of 30 journals and the proceedings of major
conferences;
Types of interventions
6. weekly current awareness alerts for a further 44 journals
Any intervention aiming to promote the initiation of breastfeed- plus monthly BioMed Central email alerts.
ing, which takes place before the first breastfeed. Evaluations of in- Search results are screened by two people and the full-text of all
terventions taking place after the first breastfeed or whose primary relevant trial reports identified through the searching activities de-
purpose is to affect the duration or exclusivity of breastfeeding are scribed above is reviewed. Based on the intervention described,
excluded from this review. each trial report is assigned a number that corresponds to a spe-
cific Pregnancy and Childbirth review topic (or topics), and is
then added to the Register. The Information Specialist searches
Types of outcome measures the Register for each review using this topic number rather than
keywords. This results in a more specific search set which has
This review includes studies that do and do not contribute out-
been fully accounted for in the relevant review sections (Included
come data.
studies; Excluded studies; Studies awaiting classification; Ongoing
studies).
See: Dyson 2005 and Fairbank 1999 for details of searching carried
Primary outcomes
out in the previous versions of this review.
1. Initiation of breastfeeding.
2. Early initiation of breastfeeding (within one hour after
birth). Searching other resources
We scanned reference lists of all relevant papers retrieved.
We did not apply any language or date restrictions.
Secondary outcomes
There were no secondary outcomes included in this review.
Data collection and analysis

For methods used in the previous versions of this review, see Dyson
Search methods for identification of studies 2005 and Fairbank 1999.
The following methods section of this review is based on a standard For this update we used the following methods when assessing the
template used by Cochrane Pregnancy and Childbirth. reports identified by the updated search.

Selection of studies (3.1) Blinding of participants and personnel (checking for
In this update, two review authors (CDG, OOB) independently possible performance bias)
assessed for inclusion all the potential studies we identified as a We described for each included study the methods used, if any, to
result of the search strategy. We resolved any disagreement through blind study participants and personnel from knowledge of which
discussion or, if required, we consulted a third person (SM). intervention a participant received. We considered that studies are
at low risk of bias if they were blinded, or if we judged that the
lack of blinding would be unlikely to affect results. We assessed
Data extraction and management
blinding separately for different outcomes or classes of outcomes.
We designed a form to extract data. For eligible studies, four re- We assessed the methods as:
view authors (EJOS, CDG, OOB, EO) extracted the data using • low, high, or unclear risk of bias for participants;
the agreed form. For studies published in abstract form only, we • low, high, or unclear risk of bias for personnel.
attempted to find full reports where available, or contacted authors
to provide same. We excluded abstracts for which full reports could
not be found. We resolved discrepancies through discussion or, if (3.2) Blinding of outcome assessment (checking for possible
required, we consulted SM. EJOS entered the data into Review detection bias)
Manager 5 software and checked for accuracy (RevMan 2014). We described for each included study the methods used, if any, to
When information regarding any of the above was unclear, we blind outcome assessors from knowledge of which intervention a
attempted to contact authors of the original reports to provide participant received. We assessed blinding separately for different
further details. outcomes or classes of outcomes.
We assessed methods used to blind outcome assessment as:
• low, high, or unclear risk of bias.
Assessment of risk of bias in included studies
Four review authors independently assessed risk of bias for each
(4) Incomplete outcome data (checking for possible attrition
study using the criteria outlined in the Cochrane Handbook for
bias due to the amount, nature, and handling of incomplete
Systematic Reviews of Interventions (Higgins 2011). We resolved
outcome data)
any disagreement by discussion.
We described for each included study, and for each outcome or
class of outcomes, the completeness of data, including attrition
(1) Random sequence generation (checking for possible and exclusions from the analysis. We stated whether attrition and
selection bias) exclusions were reported and the numbers included in the analysis
We described for each included study the method used to generate at each stage (compared with the total randomised participants),
the allocation sequence in sufficient detail to allow an assessment reasons for attrition or exclusion where reported, and whether
of whether it should produce comparable groups. missing data were balanced across groups or were related to out-
We assessed the method as: comes. Where sufficient information is reported, or can be sup-
• low risk of bias (any truly random process, e.g. random plied by the trial authors, we re-included missing data in the anal-
number table; computer random number generator); yses which we undertook.
• high risk of bias (any non-random process, e.g. odd or even We assessed methods as:
date of birth; hospital or clinic record number); • low risk of bias (e.g. no missing outcome data; missing
• unclear risk of bias. outcome data balanced across groups);
• high risk of bias (e.g. numbers or reasons for missing data
imbalanced across groups; ‘as-treated’ analysis done with
(2) Allocation concealment (checking for possible selection
substantial departure of intervention received from that assigned
bias)
at randomisation);
We described for each included study the method used to con- • unclear risk of bias.
ceal allocation to interventions prior to assignment and assessed
whether intervention allocation could have been foreseen in ad-
vance of, or during recruitment, or changed after assignment. (5) Selective reporting (checking for reporting bias)
We assessed the methods as: We described for each included study how we investigated the
• low risk of bias (e.g. telephone or central randomisation; possibility of selective outcome reporting bias and what we found.
consecutively numbered sealed opaque envelopes); We assessed the methods as:
• high risk of bias (open random allocation; unsealed or non- • low risk of bias (where it is clear that all of the study’s
opaque envelopes, alternation; date of birth); prespecified outcomes and all expected outcomes of interest to
• unclear risk of bias. the review have been reported);

• high risk of bias (where not all of the study’s prespecified Measures of treatment effect
outcomes have been reported; one or more reported primary
outcomes were not prespecified; outcomes of interest are
reported incompletely and so cannot be used; study fails to Dichotomous data
include results of a key outcome that would have been expected For dichotomous data, we presented results as summary risk ratios
to have been reported); (RRs) with 95% confidence intervals (CIs).
• unclear risk of bias.
Continuous data
(6) Other bias (checking for bias due to problems not
For continuous data, we used the mean difference if outcomes
covered by (1) to (5) above)
were measured in the same way between trials. We planned to use
We described for each included study any important concerns we the standardised mean difference to combine trials that measured
have about other possible sources of bias. the same outcome but used different methods to measure the
We assessed whether each study was free of other problems that outcome.
could put it at risk of bias:
• low risk of other bias;
• high risk of other bias; Unit of analysis issues
• unclear whether there is risk of other bias.
Cluster-randomised trials
(7) Overall risk of bias
We included cluster-randomised trials in the analyses along with
We made explicit judgements about whether studies are at high individually-randomised trials. We used the effect estimates and
risk of bias, according to the criteria given in the Cochrane Hand- uncertainty range from the cluster trials to perform the meta-
book for Systematic Reviews of Interventions (Higgins 2011). With analysis using the generic inverse variance approach for the meta-
reference to (1) to (6) above, we assessed the likely magnitude and analysis of dichotomous outcomes where trials using cluster-ran-
direction of the bias and whether we consider it is likely to im- domisation techniques were included (Alderson 2004). Further,
pact on the findings. We explored the impact of the level of bias we conducted a sensitivity analysis to investigate the effects of ran-
through undertaking sensitivity analyses - see Sensitivity analysis. domisation unit.
Assessment of the quality of the evidence using the

Other unit of analysis issues
GRADE approach
We did not include cross-over trials in this review.
For this update, we assessed the quality of the evidence using the
GRADE approach as outlined in the GRADE handbook, in order
to assess the quality of the body of evidence relating to the follow- Dealing with missing data
ing primary outcomes for the main comparisons: 1. Healthcare For included studies, we noted levels of attrition. We explored the
professional-led breastfeeding education and support versus stan- impact of including studies with high levels of missing data in the
dard care; and 2. Non-healthcare professional-led breastfeeding overall assessment of treatment effect by using sensitivity analysis.
education and support versus standard care. For all outcomes, we carried out analyses, as far as possible, on
1. Initiation of breastfeeding. an intention-to-treat basis, i.e. we attempted to include all partic-
2. Early initiation of breastfeeding (within one hour after ipants randomised to each group in the analyses, and all partici-
birth). pants were analysed in the group to which they were allocated, re-
We used GRADEpro Guideline Development Tool to import data gardless of whether or not they received the allocated intervention.
from Review Manager 5 to create ’Summary of findings’ tables The denominator for each outcome in each trial was the number
(RevMan 2014). We produced a summary of the intervention ef- randomised minus any participants whose outcomes are known
fect and a measure of quality for each of the above outcomes using to be missing.
the GRADE approach. The GRADE approach uses five consid-
erations (study limitations, consistency of effect, imprecision, in-
directness, and publication bias) to assess the quality of the body Assessment of heterogeneity
of evidence for each outcome. The evidence can be downgraded We assessed statistical heterogeneity in each meta-analysis using
from ’high quality’ by one level for serious (or by two levels for very the Tau², I² and Chi² statistics. We regarded heterogeneity as sub-
serious) limitations, depending on assessments for risk of bias, in- stantial if I² was greater than 30% and either Tau² was greater than
directness of evidence, serious inconsistency, imprecision of effect zero, or there was a low P value (less than 0.10) in the Chi² test
estimates, or potential publication bias. for heterogeneity.

Assessment of reporting biases RESULTS
In future updates, if there are 10 or more studies in a meta-analysis,
we will investigate reporting biases (such as publication bias) using
funnel plots. We will assess funnel plot asymmetry visually and Description of studies
if asymmetry is suggested by a visual assessment, we will perform This review aimed to evaluate the effectiveness of interventions
exploratory analyses to investigate it. which aim to encourage women to breastfeed in terms of changes
in the number of women who start to breastfeed and in terms of
those who initiate breastfeeding within the first hour after birth.
Data synthesis
We carried out statistical analysis using the Review Manager 5
software (RevMan 2014). We used fixed-effect meta-analysis for Results of the search
combining data where it was reasonable to assume that studies We examined 215 reports corresponding to 159 trials. We
were estimating the same underlying treatment effect, i.e. where included 28 trials and excluded 125. Four trials are ongo-
trials were examining the same intervention, and the trials’ popu- ing (Kimani-Murage 2013; NCT02084680; ISRCTN23019866;
lations and methods were judged sufficiently similar. If there was Williams 2015), and two are awaiting classification (Bakhshi 2015;
clinical heterogeneity sufficient to expect that the underlying treat- Samieizadeh 2011).
ment effects differed between trials, or if substantial statistical het-
erogeneity was detected, we used random-effects meta-analysis to Included studies
produce an overall summary if an average treatment effect across
Twenty-eight trials published between 1987 and 2016 involving
trials was considered clinically meaningful. The random-effects
107,362 women met the inclusion criteria for this review, explor-
summary was treated as the average range of possible treatment
ing the outcome of initiation of breastfeeding and early initiation
effects and we discuss the clinical implications of treatment effects
of breastfeeding (within one hour after birth). See Characteristics
differing between trials. If the average treatment effect was not
of included studies table. Outcome data was contributed by 23 tri-
clinically meaningful, we did not combine trials.
als involving 104,238 participants. Of these 23 trials, 20 were in-
Where we used random-effects analyses, the results are presented
dividually-randomised studies and three were cluster-randomised
as the average treatment effect with 95% CIs, and the estimates of
studies. Five trials met the inclusion criteria for this review but did
T² and I².
not have usable outcome data and were thus excluded from the
analyses (Caulfield 1998; Edwards 2013b; Ickovics 2007; Ickovics
Subgroup analysis and investigation of heterogeneity 2016; Sandy 2009).
When we identified substantial heterogeneity, we investigated it

using subgroup analyses and sensitivity analyses. We considered Participants
whether an overall summary was meaningful, and if it was, used Twenty of the 23 studies contributing data for the analyses and re-
random-effects analysis to produce it. porting breastfeeding initiation included a total of 27,865 partic-
We planned to carry out the following subgroup analyses for pri- ipants. For one trial (Lindenberg 1990), it was unclear how many
mary outcomes. participants were randomised to each study arm. Together, the
1. Low-income (or minority-ethnic) population versus the three cluster-randomised trials reporting early initiation of breast-
general population. feeding included 76,373 participants. Of the 20 trials reporting
We assessed subgroup differences by interaction tests available breastfeeding initiation, 14 were purposefully conducted among
within Review Manager 5 (RevMan 2014). We reported the re- low-income or deprived populations (Brent 1995; Chapman
sults of subgroup analyses quoting the χ2 statistic and P value, 2004; Chapman 2013; Coombs 1998; Edwards 2013a; Efrat
and the interaction test I² value. 2015; Hill 1987; Kellams 2016; Lindenberg 1990; MacArthur
2009; Reeder 2014; Ryser 2004; Serwint 1996; Srinivas 2015),
and six studies did not specifically target low-income or deprived
Sensitivity analysis populations (Forster 2004; Hoddinott 2009; Muirhead 2006;
We carried out sensitivity analysis to explore the effects of trial Nolan 2009; ISRCTN47056748; Wambach 2011). Three were
quality and type of randomisation on initiation of breastfeeding. conducted among specific ethnic subgroups; two recruited from
We included only trials with ’adequate’ rating on allocation con- Latina or Hispanic populations (Chapman 2004; Efrat 2015),
cealment; we considered these trials to be of high quality. We also and one recruited African-American women (Edwards 2013a).
carried out sensitivity analysis by excluding cluster-randomised Although other studies did not report that they specifically re-
trials and comparing the results of cluster-randomised trials with cruited ethnic subgroups, two trials conducted in the USA re-
the individually-randomised trials. ported that their sample comprised predominately African-Amer-

ican women (Coombs 1998; Wambach 2011). Other population 1995; Chapman 2004; Chapman 2013; Coombs 1998; Edwards
subgroups targeted by breastfeeding promotion interventions in- 2013a; Efrat 2015; Forster 2004; Hill 1987; Hoddinott 2009;
cluded women undergoing an elective, repeat caesarean section Kellams 2016; Lindenberg 1990; MacArthur 2009; Muirhead
(Nolan 2009), overweight and obese women (Chapman 2013), 2006; Nolan 2009; Reeder 2014; Ryser 2004; Serwint 1996;
and adolescents (Wambach 2011). The majority of the 20 trials ISRCTN47056748; Srinivas 2015; Wambach 2011. Three in-
reporting breastfeeding initiation were conducted among women cluded studies evaluated the effect of an intervention in terms
of mixed feeding intentions antenatally; however, three trials were of the numbers of women who initiated breastfeeding early, i.e.
conducted only among mothers who intended to breastfeed an- within one hour after birth: Flax 2014; Kirkwood 2013; Lewycka
tenatally (Chapman 2004; Chapman 2013; Reeder 2014), and 2013. Five studies did not contribute outcome data (Caulfield
one trial was conducted only among mothers who intended to 1998; Edwards 2013b; Ickovics 2007; Ickovics 2016; Sandy
formula feed or were unsure of how they intended to feed their 2009), but were included in the review.
infant (Ryser 2004).
Settings
Interventions Of the 25 trials reporting initiation of breastfeeding, 19 were
Seven trials evaluated the effect of education and support pro- conducted in the USA (Brent 1995; Caulfield 1998; Chapman
vided by non-healthcare professionals (Chapman 2004; Chapman 2004; Chapman 2013; Coombs 1998; Edwards 2013a; Edwards
2013; Edwards 2013a; Efrat 2015; MacArthur 2009; Sandy 2009; 2013b; Efrat 2015; Hill 1987; Ickovics 2007; Ickovics 2016;
Srinivas 2015), compared with standard care on breastfeeding ini- Kellams 2016; Nolan 2009; Reeder 2014; Ryser 2004; Sandy
tiation among low-income or minority-ethnic populations. Five 2009; Serwint 1996; Srinivas 2015; Wambach 2011), four
trials evaluated the effect of breastfeeding education and support were conducted in the United Kingdom of Great Britain and
compared with standard care (as defined by individual trialists) Northern Ireland (Hoddinott 2009; MacArthur 2009; Muirhead
on breastfeeding initiation (Brent 1995; Hill 1987; Ryser 2004; 2006; ISRCTN47056748), and one was conducted in Nicaragua
Serwint 1996; ISRCTN47056748). Four trials evaluated the ef- (Lindenberg 1990). Of the three trials reporting early initiation
fect of breastfeeding education using multimedia compared with of breastfeeding, one was conducted in Malawi (Lewycka 2013),
standard care on breastfeeding initiation (Caulfield 1998; Coombs one in Nigeria (Flax 2014), and one in Ghana (Kirkwood 2013).
1998; Edwards 2013b; Kellams 2016). Three trials evaluated the
effect of education and support provided by non-healthcare profes- Excluded studies
sionals compared with standard care on early initiation of breast- We excluded 125 reports from this review (see Characteristics of
feeding (Flax 2014; Kirkwood 2013; Lewycka 2013). Two trials excluded studies). Seventy-eight of these reports were not con-
evaluated the effect of education and support provided by non-
cerned with activity intended to increase breastfeeding initiation
healthcare professionals compared with standard care on breast-
rates. Thirty reports did not describe a RCT or their was insuffi-
feeding initiation among the general population (Muirhead 2006;
cient information about the study design. Eight reports described
Reeder 2014). Two trials evaluated the effect of breastfeeding ed- interventions that took place after birth, eight reports described
ucation delivered by healthcare professionals combined with peer interventions that did not target the population of interest to this
support, compared with standard care and compared with an at- review, and for one trial, breastfeeding promotion was not part
tention control intervention (Forster 2004; Wambach 2011). The
of the intervention. We excluded one trial included in the pre-
attention control intervention in Wambach 2011 was similar to the
vious version of this review and one trial awaiting classification
experimental group interventions in the amount of content and
in the previous version from this current version (Howard 2000;
timing, but did not focus on breastfeeding. Two trials evaluated the Wolfberg 2004). Howard 2000 did not involve an intervention
effect of early mother-infant contact compared with standard care for promoting breastfeeding initiation, while Wolfberg 2004 was
on breastfeeding initiation (Lindenberg 1990; Nolan 2009). Two a breastfeeding promotion intervention targeted at fathers.
trials evaluated the effect of group-based care to individualised care
(Ickovics 2007; Ickovics 2016). One trial evaluated the effect of
additional community-based breastfeeding support groups com-
pared with no additional community-based breastfeeding support
Risk of bias in included studies
groups on the rate of breastfeeding initiation (Hoddinott 2009). We conducted an assessment of studies for potential sources of
selection, performance, attrition and detection bias, and overall
risk of bias (as recommended by Higgins 2011) are detailed in
Outcomes Characteristics of included studies.
Twenty trials evaluated the effect of an intervention in terms of See Figure 1 and Figure 2 for a summary of ’Risk of bias’ assess-
the numbers of women who ever initiated breastfeeding: Brent ments.

Figure 1. ’Risk of bias’ graph: review authors’ judgements about each risk of bias item presented as
percentages across all included studies.

Figure 2. ’Risk of bias’ summary: review authors’ judgements about each risk of bias item for each included
study.

Allocation
Other potential sources of bias
Any other concerns are noted in the Characteristics of included
Sequence generation studies tables that include information about the judgements made
on the risk of bias. We assessed six studies to be at high risk of bias
In terms of methods used for random sequence generation, we
from other sources, mainly due to differences in baseline charac-
assessed over 65% (n = 19) of the 28 studies as having low risk
teristics between experimental and control groups. In two stud-
of bias, while the risk of bias was unclear for the remaining nine
ies (Efrat 2015; Ryser 2004), these differences related to infant-
studies.
feeding intentions. We judged 11 studies to have low risk of bias
from other sources, while 11 had unclear risk of bias from other
Allocation concealment sources.
We only judged eight of the 28 included studies as adequately
concealing allocation to treatment groups and therefore we con- Effects of interventions
sidered them to be at low risk of bias; we assessed 20 as having See: Summary of findings for the main comparison
unclear allocation concealment. Healthcare professional-led breastfeeding education and support
versus standard care; Summary of findings 2 Non-healthcare
Blinding professional-led breastfeeding education and support versus
standard care
Performance blinding for this type of intervention is problematic
Statistical analyses for the primary outcomes of initiation of breast-
as the women receiving the interventions and the staff delivering
feeding and early initiation of breastfeeding (within one hour after
them are likely to have been aware of group allocation. Conse-
birth) are reported below for 23 trials involving 104,238 women.
quently we did not assess any studies as being of low risk of per-
We analysed studies within seven comparisons, including Analysis
formance bias; we deemed six studies to be at unclear risk and 22
1.1, Analysis 2.1, Analysis 2.2, Analysis 3.1, Analysis 4.1, Analysis
studies at high risk.
5.1, Analysis 6.1 and Analysis 7.1.
In the case of detection bias, the objective nature of the outcome
See Summary of findings for the main comparison and Summary
being assessed, namely, whether a woman starts to breastfeed or
of findings 2 for each of the main comparisons.
not at a predefined time point, limits the scope for potential influ-
ence by the assessor, regardless of their being blind to the partici-
pant’s group allocation. However, response bias is possible where 1. Healthcare professional-led breastfeeding
outcomes are self-reported. We deemed eight studies to be of low education and support versus standard care
risk of bias, 12 were unclear and eight studies had high risk of
detection bias. The trials involving breastfeeding education delivered by health-
care professionals included the following interventions: breastfeed-
ing education and support provided during the prenatal and post-
Incomplete outcome data partum periods (Brent 1995; ISRCTN47056748); a breastfeeding
In seven studies there was significant loss to follow-up of more than lecture, including questions and answers (Hill 1987); breastfeed-
20%, or the groups were not balanced or an ’as-treated’ analysis ing promotion campaigns (Ryser 2004); and counselling (Serwint
was done leading to assessments of high risk of attrition bias. We 1996). Breastfeeding education was provided in formal settings.
assessed 16 of the studies to be of low risk of attrition bias and five Initiation of breastfeeding: there was evidence for improved breast-
studies to be of unclear risk of attrition bias. feeding initiation among women who received interventions from
healthcare professionals (average risk ratio (RR) 1.43, 95% confi-
dence interval (CI) 1.07 to 1.92; 5 trials, 564 women; Tau² = 0.07,
Selective reporting I² = 62%; Analysis 1.1; low-quality evidence). Studies included in
For most of the studies we did not have access to either trial reg- this analysis did not report early initiation of breastfeeding.
istration or the study protocol from which we could judge selec-
tive reporting. This resulted in an unclear risk of bias for selective
reporting in nearly 65% of studies (n = 18). Of the remaining 10 2. Non-healthcare professional-led breastfeeding
studies for which we had information about a priori outcomes, we education and support versus standard care
assessed six as having low risk of reporting bias and four as having The trials involving breastfeeding education delivered by non-
high risk of reporting bias. healthcare professionals included the following interventions: peer

support services provided in addition to routine care (Chapman 5712 women; Tau² = 0.02, I² = 86%; Analysis 2.1; low-quality
2004; MacArthur 2009; Muirhead 2006); peer counselling ( evidence). We found considerable heterogeneity in this analysis
Lewycka 2013; Reeder 2014; Srinivas 2015); specialised breast- and conducted a subgroup analysis based on low-income/minor-
feeding peer counselling (Chapman 2013); services from para- ity population and general population. There was no evidence of
professional doulas (Edwards 2013a); lactation educators (trained a differential effect of the interventions based on low-income/mi-
research assistants) who implemented phone-based breastfeeding nority population or the general population (test for subgroup dif-
education and support (Efrat 2015); trained credit officers who ferences: Chi² = 0.14, df = 1 (P = 0.71), I² = 0%). We conducted
led monthly breastfeeding sessions (Flax 2014); and home visits by sensitivity analysis by excluding studies with high attrition bias
community-based surveillance volunteers during pregnancy and (Chapman 2004; Chapman 2013; Efrat 2015). The overall direc-
in the first week of life (Kirkwood 2013). tion of the effect remained unchanged in favour of non-healthcare
Initiation of breastfeeding: There was evidence for improved professional-led breastfeeding education and support. Addition-
breastfeeding initiation among women who received interventions ally, statistical heterogeneity was no longer present when we ex-
delivered by non-healthcare professional counsellors and in sup- cluded studies with high attrition bias from the analysis (average
port groups (average RR 1.22, 95% CI 1.06 to 1.40; 8 trials, RR 1.23, 95% CI 1.06 to 1.43; 8 trials, 5712 women; Figure 3)
Figure 3. Sensitivity analysis (excluding high attrition bias studies) of forest plot of comparison: 2 Non-
healthcare professional-led breastfeeding education and support versus standard care, outcome: 2.1 Initiation
of breastfeeding.
Early initiation of breastfeeding: Three studies evaluated the ef-

fect of non-healthcare professional-led breastfeeding education 3. Healthcare professional-led breastfeeding
on early initiation of breastfeeding (Flax 2014; Kirkwood 2013; education with non-healthcare professional support
Lewycka 2013). When all three trials were included in the meta- versus standard care
analysis, there was a positive but non-statistically significant in-
crease in the number of women practicing early initiation of breast- Two trials involved both breastfeeding education delivered by
feeding (average RR 1.70, 95% CI 0.98 to 2.95; 3 trials, 76,373 healthcare professionals and peer support provided to mothers
women; Tau² = 0.18, I² = 78%; Analysis 2.2; low-quality evi- (Forster 2004; Wambach 2011). Wambach 2011 involved a The-
dence). We observed considerable heterogeneity in this analysis. ory of Planned Behaviour-based education and counselling inter-
vention delivered by a lactation consultant (registered nurse)-peer
counsellor team. The interventions were compared to standard
care or breastfeeding education delivered by healthcare profession- Initiation of breastfeeding: There was no evidence for improved
als not focused on breastfeeding (attention control) - see below breastfeeding initiation among women following breastfeeding ed-
4. Healthcare professional-led breastfeeding education with peer ucation interventions using multimedia (average RR 1.16, 95%
support versus attention control). CI 0.63 to 2.41; 2 trials, 497 women; Tau² = 0.18, I² = 93%;
Initiation of breastfeeding: In both trials randomising 1371 moth- Analysis 5.1). We found considerable heterogeneity between the
ers (with data available for 895 women in analysis) (Forster 2004; two studies included in this analysis. Studies included in this anal-
Wambach 2011), there was no evidence of any effect on the ini- ysis did not report early initiation of breastfeeding.
tiation of breastfeeding among mothers for breastfeeding educa-
tion delivered by healthcare professionals with peer support versus
standard care (average RR 1.06, 95% CI 0.88 to 1.27; 2 trials, 895 6. Early mother-infant contact versus standard care
women; Analysis 3.1). This study did not report early initiation Two trials that promoted mother-infant contact following ei-
of breastfeeding. ther vaginal or caesarean delivery were included in this analysis
(Lindenberg 1990; Nolan 2009). In both studies, women who
received the intervention were compared with the control group.
4. Healthcare professional-led breastfeeding Initiation of breastfeeding: There was no evidence for improved
education with peer support versus attention control breastfeeding initiation among women with increased mother-
infant contact compared to women who received usual care (RR
In one study involving 390 adolescent mothers (with data available
1.08, 95% CI 0.97 to 1.20; 2 trials, 309 women; Analysis 6.1).
for 237 women) (Wambach 2011), there was no evidence of any
Studies included in this analysis did not report early initiation of
effect on the initiation of breastfeeding among adolescent mothers
breastfeeding.
for breastfeeding education delivered by healthcare professionals
with peer support versus attention control (RR 1.21, 95% CI 0.97
to 1.51; 1 trial, 237 women; Analysis 4.1). This study did not 7. Community-based breastfeeding groups versus no
report early initiation of breastfeeding. breastfeeding groups
One trial on community-based breastfeeding groups increased the
number of breastfeeding groups available to pregnant and breast-
5. Breastfeeding education using multimedia versus feeding women in intervention localities and compared these to
routine care control localities who did not change the number of breastfeeding
Two trials involving the use of multimedia to provide breastfeed- support groups available to pregnant and breastfeeding women
ing education included the following interventions: the use of a (Hoddinott 2009). They found no difference in rates of any breast-
self-help manual seven weeks before delivery designed to commu- feeding at birth in the intervention clusters compared to the con-
nicate simple breastfeeding skills to pregnant women compared to trol clusters (mean difference (MD) -0.01, 95% CI -0.05 to 0.03;
usual breastfeeding instructions (Coombs 1998); and a low-cost 1 trial, 18,603 women; Analysis 7.1). The trialists adjusted the
breastfeeding education video shown to women prenatally versus data for pre-intervention breastfeeding rates and also for cluster-
control (Kellams 2016). ing. This trial did not report early initiation of breastfeeding.

Interventions for promoting the initiation of breastfeeding (Review) A D D I T I O N A L S U M M A R Y O F F I N D I N G S [Explanation]
Population: wom en exposed to interventions intended to prom ote breastf eeding

Setting: USA, UK, Nigeria, South Af rica, M alawi
Intervention: non-healthcare prof essional-led breastf eeding education and support
Comparison: standard care
Outcomes Anticipated absolute effects∗ (95% CI) Relative effect of participants Quality of the evidence Comments
(95% CI) (studies) (GRADE)
Risk with standard care Risk with non- health-

care professional- led
breastfeeding educa-
tion and support
Initiation of breastf eed- Study population average RR 1.22 5712 ⊕⊕ It is not possible to
ing (1.06 to 1.40) (8 RCTs) LOW 1,2 blind this type of in-
Early initiation of Study population average RR 1.70 76,373 ⊕⊕ It is not possible to

breastf eeding (0.98 to 2.95) (3 RCTs) VERY LOW 1,2,3 blind this type of in-
* The risk in the intervention group (and its 95% conf idence interval) is based on the assum ed risk in the com parison group and the relative effect of the intervention (and its
95% CI).
CI: conf idence interval; RCT: random ised controlled trial; RR: risk ratio.
GRADE Working Group grades of evidence

High quality: We are very conf ident that the true ef f ect lies close to that of the estim ate of the ef f ect.
M oderate quality: We are m oderately conf ident in the ef f ect estim ate: The true ef f ect is likely to be close to the estim ate of the ef f ect, but there is a possibility that it is
substantially dif f erent.
Low quality: Our conf idence in the ef f ect estim ate is lim ited: The true ef f ect m ay be substantially dif f erent f rom the estim ate of the ef f ect.
Very low quality: We have very little conf idence in the ef f ect estim ate: The true ef f ect is likely to be substantially dif f erent f rom the estim ate of ef f ect
17
Interventions for promoting the initiation of breastfeeding (Review)
1 M ost studies were unclear f or allocation concealm ent and som e studies were of high risk f or attrition bias. Downgraded f or
risk of bias (-1).
2
High heterogeneity I² > 80% (-1).
3 Wide 95% CI crossing the line with no ef f ect (-1).
xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx
18
DISCUSSION any point. However, it should be noted that there was substantial
heterogeneity (i.e. differences between the studies in terms of ei-
Summary of main results ther intervention, population, study design, or outcomes) which
may be a result of differences in intervention components or the
This updated review considered the evidence of the effect of in-
characteristics of the participants. For instance, four studies eval-
terventions aimed to promote the initiation of breastfeeding, tak-
uated programmes delivered in the USA to low-income women
ing place before the first breastfeed. The review includes 28 stud-
with a range of feeding intentions and where baseline breastfeeding
ies published from 1987 to 2016. In total, 107,362 women from
rates are typically low (Brent 1995; Hill 1987; Ryser 2004; Serwint
seven countries participated in the studies included in this review.
1996). Despite variation in programme components, all forms of
The majority of studies were conducted in high-income countries;
health education delivered by healthcare professionals appeared to
specifically, Australia (1 study), the USA (19 studies), and the UK
have beneficial effects in terms of breastfeeding initiation.
(4 studies), although many of these studies did specifically target
The eight studies (containing 5712 women) that we combined
low-income populations. Three studies were conducted in lower
in a meta-analysis to evaluate the effect of education and support
middle-income countries (Ghana, Nicaragua, and Nigeria), and
delivered by non-healthcare professionals provide evidence for a
one study was conducted in a low-income country (Malawi). Al-
modest improvement of breastfeeding initiation at any time point.
though the majority of studies were conducted in high-income
Again, there was a high level of heterogeneity in this analysis.
populations, only 25% of the 107,362 women included in the
Six of these studies were conducted in low-income populations,
review were from high-income countries as the study from Malawi
which may call into question the generalisability of the results.
was very large, with 55,931 participants (Lewycka 2013).
However, when we compared studies of general populations with
All of the studies conducted in high-income settings and the study
low-income populations, we did not find any differences.
conducted in Nicargua evaluated whether the intervention had an
When we combined the three studies (containing 76,373 women)
effect on the number of women who ever initiated breastfeeding.
that evaluated the effect of non-healthcare professional support on
Only three of the 28 studies (the studies conducted in the other
early breastfeeding initiation, we did not identify any evidence of
non-high income countries) evaluated whether the intervention
an effect. This analysis also suggested a high level of heterogeneity.
had any effect on the number of women who initiated breastfeed-
When we excluded Flax 2014 from the analysis, a trial with possible
ing early (i.e. within one hour of birth).
selection bias, there was a statistically significant increase in the
Of those studies contributing data, the nature of the intervention
number of women who practiced early initiation of breastfeeding
varied between studies. Specifically, five studies evaluated the ef-
and no evidence of heterogeneity.
fect of breastfeeding education and support provided by health-
Two studies examined the effect of combined healthcare pro-
care professionals. Eleven studies evaluated the effect of education
fessional-led education with telephone peer support (Wambach
and support provided by non-healthcare professionals (i.e. peer/lay
2011), or community educator (Forster 2004). The intervention
support). Of these, nine were conducted in low-income or ethnic
for Wambach 2011 was specifically targeted at adolescent moth-
minority populations, and two were conducted in the general pop-
ers (n = 390) and did not demonstrate any effect on breastfeed-
ulation. Two studies examined combined healthcare professional
ing initiation. Similarly, there was no evidence of an intervention
and non-healthcare professional support. Two trials examined the
effect in the studies which looked at multimedia based interven-
effect of multimedia breastfeeding education programmes, and
tions. This included one trial of a self-help manual of 200 women
another two trials examined the effect of early mother-infant con-
(Coombs 1998), and one trial of a breastfeeding video of 522
tact. Finally, one trial examined the effect of additional commu-
women (Kellams 2016).
nity-based breastfeeding support groups. It should also be noted
The two studies that examined the effect of early mother-infant
that even within the same intervention type, the actual compo-
contact immediately after vaginal birth (Lindenberg 1990), and
nents of the intervention also varied. In particular, delivery of
following caesarean section (Nolan 2009), showed no evidence
the non-healthcare professional education and support included:
of effect on breastfeeding initiation specifically. However, the lit-
education and support provided by peer supporter/counsellors
erature on the promotion of the duration of breastfeeding pro-
(Chapman 2004; Chapman 2013; Lewycka 2013; MacArthur
vides clear evidence of the benefits of ongoing mother and infant
2009; Muirhead 2006; Reeder 2014; Srinivas 2015), para-profes-
contact during the hospital stay to support the mother’s ability to
sional doulas (Edwards 2013a), trained research assistants (Efrat
breastfeed (Moore 2012).
2015), trained credit officers (Flax 2014), and community-based
Finally, Hoddinott 2009 was the one study that examined the ef-
surveillance volunteers (Kirkwood 2013).
fect of providing new, additional community-based breastfeeding
The pooled data of the five studies (containing 564 women) ex-
support groups in low-income areas compared to existing breast-
amining the effect of health education interventions delivered by
feeding groups; it found no effect on breastfeeding initiation.
healthcare professionals indicated that health education interven-
tions delivered by healthcare professionals had a modest effect on
increasing the number of women who initiated breastfeeding at

Overall completeness and applicability of ering and evaluating breastfeeding promotion interventions, the
evidence ability to effectively blind participants and personnel and thereby
reduce performance bias is limited. It is therefore unsurprising that
This updated review now contains 28 studies, out of which 23
we assessed all 28 studies as having high or unclear risk of perfor-
studies contribute data to the review. The number of women in-
mance bias. This should be recognised as an inherent weakness
cluded in the review has increased considerably from 1553 in the
of this particular type of evidence base, rather than of the partic-
previous version of the review (Dyson 2005), to 107,362 in this
ular studies included in this review. Of more concern is that we
update; this in part is driven by a very large study conducted in
assessed 20 studies to have high or unclear risk of detection bias.
Malawi, which included 55,931 women (Lewycka 2013). There-
Even where there was blinding of outcome assessment, there is a
fore, whilst the majority of studies in this review were conducted
risk of response bias in self-reported outcomes where participants
in high-income countries (Australia, USA, and UK), the majority
were not blinded.
of participants were from low- and middle-income countries (over
Incomplete outcome data was also a source of possible bias in this
70% of all women). The studies in the USA tended to be small
review as we assessed only 16 of the studies as having low risk of
and only involved a total of 5566 women. They also generally
attrition bias. The remaining 12 studies either had high rates of loss
targeted specific low-income or disadvantaged groups. The high
to follow-up or failed to report attrition clearly. To minimise the
preponderance of trials from the USA raises questions about the
effect of this, we conducted all analyses on the basis of intention-
applicability of the findings to other settings.
to-treat. However, it is possible that this approach may dilute the
The interventions tested across the studies included in this re-
actual effect of the interventions.
view were very diverse. For example, the educational interventions
We only assessed six studies as being at low risk of bias for selective
delivered by healthcare professionals included several distinct ap-
outcome reporting; we judged four at high risk and 18 at unclear
proaches: a series of one-to-one sessions with a lactation consul-
risk. The high number judged as unclear risk was due to the lack
tation (Brent 1995); a 40-minute lecture with time for questions
of protocols or trial registration detailing prespecified outcomes.
(Hill 1987); and one session with a paediatrician that covered
We assessed six studies as having a high risk of bias from other
a range of topics, of which breastfeeding was just one (Serwint
sources, including differences in baseline characteristics (Brent
1996). Standard care was also diverse across the included studies
1995; Caulfield 1998; Efrat 2015; Ickovics 2007; Ryser 2004;
and, in the case of the UK where breastfeeding support is part of
Wambach 2011). Specifically, in the study by Efrat 2015, women
standard postnatal care, it is perhaps not surprising that some in-
in the experimental group had a significantly higher intention to
terventions did not have an effect above and beyond that of stan-
breastfeed than those in the control group. Similary, in the study
dard care. For example, in one trial (ISRCTN47056748), women
by Ryser 2004, more participants in the experimental group were
in both experimental and control groups received care that met
undecided about feeding decisions, while more participants in the
UNICEF Baby-Friendly standards and included a two-hour ante-
control group planned to formula feed. Of the remaining studies,
natal breastfeeding education class. Another trial assessed commu-
we assessed 11 to be of low risk of bias and 11 to be of unclear risk
nity-based support groups in a community where existing breast-
of bias from other sources.
feeding support groups were available for control group partici-
We assessed the quality of the evidence in this review using the
pants (Hoddinott 2009.
GRADE approach (Atkins 2004). See Summary of findings for
Caution is needed in interpreting the findings of the two trials
the main comparison and Summary of findings 2. For the com-
on early mother-infant contact (Lindenberg 1990; Nolan 2009).
parison of healthcare professional-led breastfeeding education and
Generalisation of the results is not recommended due to the mod-
support versus standard care, we assessed the quality of evidence
erate quality and size of the studies, and to fundamental concerns
for the outcome of initiation of breastfeeding as low (Summary of
regarding the practice of routine separation of mother and infant
findings for the main comparison). We downgraded the quality
prior to hospital discharge in the case of Lindenberg 1990, and
of evidence due to design limitations for most of the studies that
separation following caesarean section in Nolan 2009.
contributed data and also high statistical heterogeneity (I2 more
than 60%). For the comparison of non-healthcare professional-led
breastfeeding education and support versus standard care, there
Quality of the evidence was also low-quality evidence for the outcome of any initiation of
breastfeeding due to design limitations in trials (unclear allocation
Overall, we judged the methodological quality of the studies in-
concealment and high risk for attrition bias) and again high hetero-
cluded in this review to be mixed. While we assessed over 65% of
geneity (Summary of findings 2). We also deemed early initiation
the studies to have low risk of bias for generating randomisation
of breastfeeding to be of very low-quality of evidence; downgraded
sequence, we only judged seven studies to have adequately con-
due to lack of blinding, high heterogeneity and imprecision that
cealed group allocation. This raises concerns regarding the effect
was demonstrated with a wide 95% CI that crossed the line with
of selection bias on study findings.
no effect (Summary of findings 2).
Given that there are genuine pragmatic considerations when deliv-

Potential biases in the review process likely to result in some improvements in breastfeeding initiation
rates, particularly among low-income or minority-ethnic women
Bias can potentially be introduced at any stage of the review pro-
in the USA, where baseline breastfeeding rates are typically low.
cess. To minimise this, two review authors independently screened
Similarly, breastfeeding interventions provided by non-healthcare
studies for inclusion and any disagreements were resolved by a
professionals could lead to improvements in rates of early initia-
third review author. Data extraction and ’Risk of bias’ assessments
tion of breastfeeding in low-income countries.
were performed by one review author and then checked by a sec-
ond review author. Again, any discrepancies were resolved by a The type of education or support intervention which may be most
third review author. ’Risk of bias’ assessment is subjective in nature likely to increase initiation rates appears to be needs-based, one-
and therefore another team of review authors may have graded to-one, informal sessions delivered in the antenatal or perinatal
studies differently. To minimise language bias, we translated any period by a trained breastfeeding professional or peer counsellor.
study not reported in English into English, and included it in This review update mainly included studies conducted in the USA
the review, providing it met the inclusion criteria. Whilst we at- among low-income women, thus generalisability may be limited
tempted to identify all the evidence on interventions for the initi- to populations of similar characteristics.
ation of breastfeeding (including published abstracts from confer-
ence proceedings) and followed up ongoing studies, it is feasible Breastfeeding education using multimedia may not be an effective
that relevant research which is unpublished or not registered in a breastfeeding promotion strategy particularly among low-income
clinical trials register could have been missed. women.
Early mother-infant contact for women with vaginal or caesarean

deliveries was not effective in improving breastfeeding initiation
Agreements and disagreements with other rates. Enabling mothers and infants to remain together for 24
studies or reviews hours a day, ’rooming-in,’ is one of the Ten Steps of the UNICEF/
Consistent with this review, other reviews have reported that inter- WHO Baby Friendly Hospital Initiative (BFHI) adopted as a
ventions including (Jolly 2012; Rollins 2016; Sinha 2015): health global programme to support successful breastfeeding and demon-
education and counselling provided by healthcare professionals; strated to increase initiation rates for all women in all settings.
education provided by non-healthcare professionals; and peer sup-
port, can increase both the number of women who ever initi- Implications for research
ate breastfeeding and those who initiate breastfeeding within the
first hour after birth. Other reviews provide additional evidence The majority of the studies included in this review were con-
regarding the implementation of such interventions. For exam- ducted in the USA and the effectiveness of interventions reviewed
ple, interventions that are delivered in a combination of settings here needs to be assessed widely in diverse countries and settings,
in studies that are adequately powered, have adequate methods
(e.g. home and community, or health systems and community)
of randomisation, adequate reporting of losses to follow-up, and
are more effective than those delivered in one setting only (Sinha
2015). Additionally, Beake 2012 reported that in health system utilise intention-to-treat analysis.
settings where breastfeeding initiation rates are low, structured Publication of evaluations of effectiveness should detail the con-
programmes of interventions may be most effective (Beake 2012). tent and method of the intervention delivered; the people (e.g.
Moreover, Pérez-Escamilla 2016 found a dose-response between peer or healthcare professional) who delivered it and the training
the number of Baby Friendly Hospital Initiative (BFHI) steps and experience these people had; baseline breastfeeding rates for
women are exposed to and the likelihood of improved breastfeed- the study-site population; and feeding intention for participants
ing outcomes, including early breastfeeding initiation. The use of within each comparison group.
new technologies may also be an area for future development, with
one study in the review by Rollins 2016 suggesting that mass or Future research should aim to evaluate the effectiveness of the
social media promotion of breastfeeding potentially has a major intervention to improve both the initiation and duration of any
effect on early initiation of breastfeeding. and/or exclusive breastfeeding at least up to six months to enable
appropriate planning and implementation of interventions dur-
ing pregnancy and the postnatal period. In addition, studies need
to provide clear descriptions of both the intervention and study
AUTHORS’ CONCLUSIONS outcomes.
Further research to evaluate interventions that combine health

Implications for practice education or support before the birth with support during the days
Health education and counselling provided by healthcare profes- immediately after the birth should be evaluated and compared
sionals and peer support interventions included in this review are with those that offer education alone.

Further research into early mother-infant contact regardless of of findings for the Health Technology Assessment (HTA) 2000
mode of delivery, followed by rooming-in until hospital discharge review. Members of the Advisory Panel were: Rosamund Bryar,
is needed to evaluate the effect of early mother-infant contact on Petra Clarke, Leslie Davidson, Elisabeth Helsing, Stuart Logan,
increasing breastfeeding initiation rates among various population Miranda Mugford, Patricia Muirhead, Felicity Savage, Jim Siko-
groups. rski, and Mary Smale.
Studies are needed to help women to find ways to overcome societal Searches for studies relevant to this substantive update of this
barriers to breastfeeding, including policy-level interventions. Cochrane Review were conducted by Lynn Hampson (Informa-
tion Specialist, Cochrane Pregnancy and Childbirth, Liverpool
Good-quality research to evaluate the effectiveness of breastfeeding
Women’s Hospital Foundation Trust).
promotion and support on breastfeeding rates among maternity
and community services who achieve fully accredited BFI/BFHI Thanks to Hannah Soley for translating Lucchini 2013.
status would further inform policy and practice.
Professor Martin Bland, University of York, provided statistical
advice for a previous update.
This research was supported by a grant from the Evidence and Pro-
gramme Guidance Unit, Department of Nutrition for Health and
ACKNOWLEDGEMENTS
Development, World Health Organization (WHO). The findings,
Developmental work for the original review (development of interpretations and conclusions expressed in this paper are entirely
the conceptual framework for identification and classification of those of the authors and should not be attributed in any manner
health promotion interventions and the search strategy) was con- whatsoever to the WHO.
ducted for the purposes of ’A systematic review to evaluate the
This project was supported by the National Institute for
effectiveness of interventions to promote the initiation of breast-
Health Research (NIHR), via Cochrane Infrastructure funding
feeding’, funded by and produced for the National Health Ser-
to Cochrane Pregnancy and Childbirth. The views and opinions
vice (NHS) Research and Development Health Technology As-
expressed therein are those of the authors and do not necessarily
sessment Programme, UK (Fairbank 2000). Co-authors of that re-
reflect those of the Systematic Reviews Programme, NIHR, NHS,
view, not involved in the adaptation and update of this Cochrane
or the Department of Health.
Review were: S O’Meara, Dr AJ Sowden, D Lister-Sharp (NHS
Centre for Reviews and Dissemination, University of York, UK), As part of the pre-publication editorial process, this review has
and Dr M Woolridge (Mother and Infant Research Unit, Faculty been commented on by two peers (an editor and one referee who
of Medicine, University of Leeds, UK). An international Advi- is external to the editorial team), a member of the Pregnancy
sory Panel of breastfeeding and health promotion experts also pro- and Childbirth Group’s international panel of consumers and the
vided guidance on the classification of studies and interpretation Group’s statistical adviser.
REFERENCES
References to studies included in this review duration. Advances in Experimental Medicine and Biology
2004;554:303–6. [3414074]
Brent 1995 {published data only} Chapman DJ, Damio G, Pérez-Escamilla R. Differential
Brent NB, Redd B, Dworetz A, D’Amico F, Greenberg JJ. response to breastfeeding peer counseling within a low-
Breast-feeding in a low income population: program to income, predominantly Latina population. Journal of
increase incidence and duration. Archives of Pediatrics & Human Lactation 2004;20(4):389–96. [3414075]
Adolescent Medicine 1995;149(7):798–803. [3414070] ∗
Chapman DJ, Damio G, Young S, Perez-Escamilla R.
Caulfield 1998 {published and unpublished data} Effectiveness of breastfeeding peer counseling in a low-
Caulfield LE, Gross SM, Bentley ME, Bronner Y, income, predominantly Latina population. Archives of
Kessler L, Jensen J, et al. WIC-based interventions to Pediatrics & Adolescent Medicine 2004;158:897–902.
promote breastfeeding among African-American women [3414076]
in Baltimore: effects on breastfeeding initiation and Chapman DJ, Perez-Escamilla R. Acculturative type is
continuation. Journal of Human Lactation 1998;14(1): associated with breastfeeding duration among low-income
15–22. [3414072] Latinas. Maternal and Child Nutrition 2013;9(2):188–98.
Chapman 2004 {published data only} [4446957]
Chapman D, Damio G, Young S, Perez-Escamilla
R. Association of degree and timing of exposure to Chapman 2013 {published data only}
breastfeeding peer counseling services with breastfeeding Chapman DJ, Bermudez-Millan A, Wetzel K, Damio G,
Kyer N, Young S, et al. Breastfeeding education and support controlled trial. BMC Pregnancy and Childbirth 2003;3:5.
trial for obese women. FASEB 2008;22:1080.4. [4446958] [3414081]
∗
Chapman DJ, Morel K, Bermúdez-Millán A, Young S, Forster DA, McLachan HL, Lumley J. Factors associated
Damio G, Pérez-Escamilla R. Breastfeeding education with breastfeeding at six months postpartum in a group of
and support trial for overweight and obese women: Australian women. International Breastfeeding Journal 2006;
a randomized trial. Pediatrics 2013;131(1):e167–70. 1:18. [3414082]
[4446959] Forster DA, McLachlan HL, Lumley J. Risk factors for
early cessation of breastfeeding: results from a randomised
Coombs 1998 {published data only} controlled trial. Perinatal Society of Australia and New
Coombs DW, Reynolds K, Joyner G, Blankson M. A self- Zealand 10th Annual Congress; 2006 April 3-6; Perth,
help program to increase breastfeeding among low-income Australia. 2006:149. [3414084]
women. Journal of Nutrition Education 1998;30(4):203–9. Forster DA, McLachlan HL, Lumley J, Beanland CJ,
[3414078] Waldenstrom U, Short RV, et al. ABFAB: attachment to the
breast and family attitudes to breastfeeding. The effect of
Edwards 2013a {published data only} breastfeeding education in the middle of pregnancy on the
Edwards C, Thullen J, Korfmacher J, Lantos D, Henson
duration of breastfeeding: a randomised controlled trial
G, Hans L. Breastfeeding and complementary food: [abstract]. Perinatal Society of Australia and New Zealand
randomized trial of community doula home visiting.
7th Annual Congress; 2003 March 9-12; Tasmania,
Pediatrics 2013;132:S160–6. [3414349] Australia. 2003:A70. [3414083]
Edwards 2013b {published data only} Hill 1987 {published data only}
Edwards RA, Bickmore T, Jenkins L, Foley M, Manjourides Hill PD. Effects of education on breastfeeding success.
J. Use of an interactive computer agent to support Maternal-Child Nursing Journal 1987;16(2):145–6.
breastfeeding. Maternal & Child Health Journal 2013;17 [3414086]
(10):1961–8. [3414351]
Hoddinott 2009 {published data only}
Efrat 2015 {published data only} Hoddinott P. A randomised controlled trial to evaluate the
Efrat MW, Esparza S, Mendelson SG, Lane CJ. The effect clinical and cost effectiveness of breastfeeding peer support
of lactation educators implementing a telephone-based groups in improving breastfeeding initiation, duration
intervention among low-income Hispanics: a randomised and satisfaction. www.nrr.nhs.uk (accessed 13 Aug 2007).
trial. Health Education Journal 2015;74(4):424–41. [3414090]
[3414353] Hoddinott P, Britten J, Pill R. Why do interventions work
in some places and not others: a breastfeeding support
Flax 2014 {published data only} group trial. Social Science & Medicine 2010;70(5):769–78.
Flax V, Negerie M, Usman A, Leatherman S, Daza E, [3414088]
Bentley M. Nigerian women participating in an integrated ∗
Hoddinott P, Britten J, Prescott GJ, Tappin D, Ludbrook
microcredit and mhealth breastfeeding promotion A, Godden DJ. Effectiveness of policy to provide
intervention were more likely to adopt international breastfeeding groups (BIG) for pregnant and breastfeeding
breastfeeding recommendations. Annals of Nutrition & mothers in primary care: cluster randomised controlled
Metabolism 2013;63(Suppl 1):885, Abstract no: PO1294. trial. BMJ 2009;338:a3026. [3414089]
[4446960]
∗
Flax VL, Negerie M, Ibrahim AU, Leatherman S, Daza Ickovics 2007 {published data only (unpublished sought but not used)}
∗
EJ, Bentley ME. Integrating group counseling, cell phone Ickovics JR, Kershaw TS, Westdahl C, Magriples
messaging, and participant-generated songs and dramas U, Massey Z, Reynolds H, et al. Group prenatal care
into a microcredit program increases Nigerian women’s and perinatal outcomes: a randomized controlled trial.
adherence to international breastfeeding recommendations. Obstetrics and Gynecology 2007;110(2 Pt 1):15. [3414094]
Journal of Nutrition 2014;144(7):1120–4. [3414367] Novick G, Reid E, Lewis J, Kershaw S, Rising SS, Ickovics
R. Group prenatal care: model fidelity and outcomes.
Forster 2004 {published data only} American Journal of Obstetrics & Gynecology 2013;209(2):
∗
Forster D, McLachlan H, Lumley J, Beanland C, 112.e1–112.e6. [4446961]
Waldenstrom U, Amir L. Two mid-pregnancy interventions Novick G, Reid E, Lewis J, Kershaw T, Rising S, Ickovics
to increase the initiation and duration of breastfeeding: a R. Group prenatal care: model fidelity and outcomes.
randomized controlled trial. Birth 2004;31(3):176–82. Journal of Midwifery & Women’s Health 2013;58(5):586–7.
[3414080] [4446962]
Forster D, McLachlan H, Lumley J, Beanland C, Westdahl CM, Kershaw T, Schindler-Rising S, Ickovics J.
Waldenstrom U, Harris H, et al. ABFAB. Attachment to Group prenatal care improves breastfeeding initiation and
the breast and family attitudes to breastfeeding. The effect duration: results from a two-site randomized controlled
of breastfeeding education in the middle of pregnancy on trial. Journal of Human Lactation 2008;24(1):96–7.
the initiation and duration of breastfeeding: a randomised [4446963]
Ickovics 2016 {published data only} Nolan 2009 {published data only}
Ickovics JR, Earnshaw V, Lewis JB, Kershaw TS, Magriples Nolan A, Lawrence C. A pilot study of a nursing intervention
U, Stasko E, et al. Cluster randomized controlled trial of protocol to minimize maternal-infant separation after
group prenatal care: perinatal outcomes among adolescents Cesarean birth. Journal of Obstetric, Gynecologic, and
in New York City health centers. American Journal of Public Neonatal Nursing 2009;38(4):430–42. [3414103]
Health 2016;106(2):359–65. [3414379] Reeder 2014 {published data only}
ISRCTN47056748 {published data only} Reeder JA, Joyce T, Sibley K, Arnold D, Altindag O.
ISRCTN47056748. Successful breastfeeding promotion: Telephone peer counseling of breastfeeding among WIC
a motivational instructional model applied and tested. participants: a randomized controlled trial. Pediatrics 2014;
isrctn.com/ISRCTN47056748 (first received 16 July 2007). 134(3):e700–9. [3414431]
[3414114] Ryser 2004 {published data only}
Kellams 2016 {published data only} Ryser FG. Breastfeeding Attitudes, Intention and Initiation in
Kellams AL, Gurka KK, Hornsby PP, Drake E, Riffon M, Low-income Women: the Effect of the “Best Start” Program.
Gellerson D, et al. The impact of a prenatal education video Texas: Texas Woman’s University, 1999. [3414105]
on rates of breastfeeding initiation and exclusivity during ∗
Ryser FG. Breastfeeding attitudes, intention, and initiation
the newborn hospital stay in a low-income population. in low-income women: the effect of the best start program.
Journal of Human Lactation 2016;32:152–9. [3414389] Journal of Human Lactation 2004;20(3):300–5. [3414106]
Kirkwood 2013 {published data only} Sandy 2009 {published and unpublished data}
Kirkwood BR, Manu A, ten Asbroek AH, Soremekun S, Sandy JM, Anisfeld E, Ramirez E. Effects of a prenatal
Weobong B, Gyan T, et al. Effect of the Newhints home- intervention on breastfeeding initiation rates in a Latina
visits intervention on neonatal mortality rate and care immigrant sample. Journal of Human Lactation 2009;25(4):
practices in Ghana: a cluster randomised controlled trial. 404–11. [3414108]
Lancet 2013;381(9884):2184–92. [3414393] Serwint 1996 {published data only}
Lewycka 2013 {published data only} Serwint JR, Wilson MEH, Vogelhut JW, Repke JT, Seidel
Lewycka S, Mwansambo C, Kazembe P, Phiri T, Mganga HM. A randomized controlled trial of prenatal pediatric
A, Rosato M, et al. A cluster randomised controlled visits for urban, low-income families. Pediatrics 1996;98(6):
trial of the community effectiveness of two interventions 1969–75. [3414112]
in rural Malawi to improve health care and to reduce Srinivas 2015 {published data only}
maternal, newborn and infant mortality. Trials 2010;11:88. ∗
Srinivas GL, Benson M, Worley S, Schulte E. A clinic-
[4446964] based breastfeeding peer counselor intervention in an urban,
∗
Lewycka S, Mwansambo C, Rosato M, Kazembe P, Phiri low-income population: Interaction with breastfeeding
T, Mganga A, et al. Effect of women’s groups and volunteer attitude. Journal of Human Lactation 2015;31(1):120–8.
peer counselling on rates of mortality, morbidity, and [3414443]
health behaviours in mothers and children in rural Malawi Srinivas GL, Worley S. Effect of office-based peer counselor
(MaiMwana): a factorial, cluster-randomised controlled on breastfeeding rates in an urban low-income clinic.
trial. Lancet 2013;381(9879):1721–35. [3414399] Pediatric Academic Societies Annual Meeting; 2013 May 4-
Lindenberg 1990 {published data only} 7; Washington DC, USA. 2013. [4446965]
Lindenberg CS, Artola RC, Jimenez V. The effect of early Wambach 2011 {published data only}
post-partum mother-infant contact and breast-feeding Wambach K. Kansas University Teen Mothers Project.
promotion on the incidence and continuation of breast- clinicaltrials.gov/ct2/show/NCT00222118 (first received
feeding. International Journal of Nursing Studies 1990;27 13 September 2005). [3414117]
(3):179–86. [3414096] Wambach K, Rojjanasrirat W, Williams Domian E,
MacArthur 2009 {published data only} Aaronson L, Breedlove G, Yeh HW. Effects of a peer
Jolly K, Ingram L, Freemantle N, Khan K, Chambers counselor and lactation consultant on breastfeeding
J, Hamburger R, et al. Effect of a peer support service initiation and duration. Journal of Human Lactation 2009;
on breast-feeding continuation in the UK: a randomised 25(1):101–2. [3414116]
∗
controlled trial. Midwifery 2012;28(6):740–5. [3414098] Wambach KA, Aaronson L, Breedlove G, Domian EW,
∗ Rojjanasrirat W, Yeh HW. A randomized controlled trial of
MacArthur C, Jolly K, Ingram L, Freemantle N, Dennis
CL, Hamburger R, et al. Antenatal peer support workers breastfeeding support and education for adolescent mothers.
and initiation of breast feeding: cluster randomised Western Journal of Nursing Research 2011;33(4):486–505.
controlled trial. BMJ 2009;338:b131. [3414099] [3414118]
Muirhead 2006 {published data only} References to studies excluded from this review
Muirhead P. The effect of a programme of organised and
supervised peer support on the initiation and duration of Ahmad 2012 {published data only}
breastfeeding: a randomised trial. British Journal of General Ahmad MO, Sughra U, Kalsoom U, Imran M, Hadi U.
Practice 2006;56:191–7. [3414101] Effect of antenatal counselling on exclusive breastfeeding.
Journal of Ayub Medical College, Abbottabad: JAMC 2012; Obstetric, Gynecologic and Neonatal Nursing 1978;13:54.
24(2):116–9. [3414325] [3414133]
Ahmed 2008 {published data only} Bonuck 2005 {published data only}
Ahmed AH. Breastfeeding preterm infants: an educational Bonuck KA, Freeman K, Trombley M. Randomized
program to support mothers of preterm infants in Cairo, controlled trial of a prenatal and postnatal lactation
Egypt. Pediatric Nursing 2008;34(2):125–30. [3414122] consultant intervention on infant health care use. Archives
Aidam 2005 {published data only} of Pediatrics & Adolescent Medicine 2006;160(9):953–60.
Aidam BA, Perez-Escamilla R, Lartey A. Lactation [3414135]
∗
counseling increases exclusive breast-feeding rates in Ghana. Bonuck KA, Trombley M, Freeman K, McKee D.
Journal of Nutrition 2005;135(7):1691–5. [3414124] Randomized, controlled trial of a prenatal and postnatal
lactation consultant intervention on duration and intensity
Anderson 2005 {published data only} of breastfeeding up to 12 months. Pediatrics 2005;116(6):
Anderson AK, Damio G, Young S, Chapman DJ, Perez- 1413–26. [3414136]
Escamilla R. A randomized trial assessing the efficacy of peer Memmott MM, Bonuck KA. Mother’s reactions to a skills-
counseling on exclusive breastfeeding in a predominantly based breastfeeding promotion intervention. Maternal and
Latina low-income community. Archives of Pediatrics & Child Nutrition 2006;2(1):40–50. [3414137]
Adolescent Medicine 2005;159(9):836–41. [3414126]
Bonuck 2013 {published data only}
Andersson 2013 {published data only}
Bonuck K, Stuebe A, Barnett J, Fletcher J, Bernstein P.
Andersson E, Christensson K, Hildingsson I. Mothers’ Routine, primary-care based interventions to increase
satisfaction with group antenatal care versus individual
breastfeeding: results of two randomized controlled trials.
antenatal care - A clinical trial. Sexual and Reproductive Breastfeeding Medicine 2013;8(Suppl 1):S–19. [3414337]
Healthcare 2013;4(3):113–20. [3414327]
Bottaro 2009 {published data only}
Babakazo 2015 {published data only}
Bottaro SM, Giugliani ER. Effectiveness of an intervention
Babakazo P, Donnen P, Mapatano MA, Lulebo A,
to improve breastfeeding knowledge and attitudes among
Okitolonda E. Effect of the baby friendly hospital initiative
fifth-grade children in Brazil. Journal of Human Lactation
on the duration of exclusive breastfeeding in Kinshasa: a
2009;25(3):325–32. [3414139]
cluster randomized trial. Revue D’epidemiologie Et De Sante
Publique 2015;63:285–92. [3414329] Byrne 2000 {published data only}
Ball 2006 {published data only} Byrne JP, Crowther CA, Moss JR. A randomised controlled
Ball HL, Ward-Platt MP, Heslop E, Leech SJ, Brown KA. trial comparing birthing centre care with delivery suite
Randomised trial of infant sleep location on the postnatal care in Adelaide, Australia. Australian and New Zealand
ward. Archives of Disease in Childhood 2006;91:1005–10. Journal of Obstetrics and Gynaecology 2000;40(3):268–74.
[3414128] [3414141]
Ball 2011 {published data only} Carfoot 2001 [pers comm] {published and unpublished data}
Ball H. Sleeping and feeding in the first 6 months: test of a Carfoot S. Successful breastfeeding: the effects of skin-to-
large-scale data collection technique. Journal of Reproductive skin [personal communication]. Personal communication
and Infant Psychology 2004;22(3):231. [3414130] 2001. [3414143]
∗
Ball HL, Ward-Platt MP, Howel D, Russell C. Randomised Carfoot 2005 {published data only}
trial of sidecar crib use on breastfeeding duration ∗
Carfoot S, Williamson P, Dickson R. A randomised
(NECOT). Archives of Disease in Childhood 2011;96(7): controlled trial in the north of England examining the
630–4. [3414131] effects of skin-to-skin care on breastfeeding. Midwifery
Begley 2011 {published data only} 2005;21:71–9. [3414145]
Begley C, Devane D, Clarke M, McCann C, Hughes P, Carfoot S, Williamson PR, Dickson R. The value of a
Reilly M, et al. Comparison of midwife-led and consultant- pilot study in breast-feeding research. Midwifery 2004;20:
led care of healthy women at low risk of childbirth 188–93. [3414146]
complications in the Republic of Ireland: a randomised trial.
Cattaneo 2001 {published data only}
BMC Pregnancy and Childbirth 2011;11:85. [3414068] Cattaneo A, Buzzetti R. Effect on rates of breast feeding of
Bica 2014 {published data only} training for the baby friendly hospital initiative. BMJ 2001;
Bica OC, Giugliani ER. Influence of counseling sessions 323(7325):1358–62. [3414148]
on the prevalence of breastfeeding in the first year of life:
Chapman 1986 {published data only}
a randomized clinical trial with adolescent mothers and
Chapman MG, Jones M, Spring JE, de Swiet M,
grandmothers. Birth 2014;41(1):39–45. [3414333]
Chamberlain GV. The use of a birthroom: a randomized
Bishop 1978 {unpublished data only} controlled trial comparing delivery with that in the labour
Bishop WS. An educational program on breastfeeding for ward. British Journal of Obstetrics and Gynaecology 1986;98:
maternity nurses. A cost-effective evaluation. Journal of 182–7. [3414150]
Chapman 2011 {published data only} income women: a preliminary study. Cadernos de Saude
Chapman DJ, Morel K, Bermudez-Millan A, Young S, Publica 1996;12(1):61–8. [3414168]
Damio G, Kyer N, et al. Breastfeeding education and Garmendia 2015 {published data only}
support trial for obese women: effects of a specialized Garmendia ML, Corvalan C, Araya M, Casanello P,
peer counseling intervention on breastfeeding and health Kusanovic JP, Uauy R. Effectiveness of a normative nutrition
outcomes. Journal of Human Lactation 2011;27(1):75–6. intervention (diet, physical activity and breastfeeding) on
[3414152] maternal nutrition and offspring growth: the Chilean
Coutinho 2005 {published data only} maternal and infant nutrition cohort study (CHiMINCs).
Coutinho SB, de Lira PI, de Carvalho Lima M, Ashworth BMC Pregnancy and Childbirth 2015;15:175. [3414369]
A. Comparison of the effect of two systems for the Girish 2013 {published data only}
promotion of exclusive breastfeeding. Lancet 2005;366 Girish M, Mujawar N, Gotmare P, Paul N, Punia S, Pandey
(9491):1094–100. [3414156] P. Impact and feasibility of breast crawl in a tertiary care
Di Napoli 2004 {published data only} hospital. Journal of Perinatology 2013;33(4):288–91.
Di Napoli A, Di Lallo D, Fortes C, Franceschelli C, Armeni [3414371]
E, Guasticchi G. Home breastfeeding support by health Gordon 1999 {published data only}
professionals: findings of a randomized controlled trial in a Gordon NP, Walton D, McAdam E, Derman J, Gallitero
population of Italian women. Acta Paediatrica 2004;93(8): G, Garrett L. Effects of providing hospital-based doulas in
1108–14. [3414158] health maintenance organization hospitals. Obstetrics &
Doherty 2012 {published data only} Gynecology 1999;93(3):422–6. [3414170]
Doherty T, Sanders D, Jackson D, Swanevelder S, Lombard Graffy 2001 [pers comm] {published and unpublished data}
C, Zembe W, et al. Early cessation of breastfeeding amongst Graffy J. A randomised controlled trial of the effectiveness
women in South Africa: an area needing urgent attention of support from breastfeeding counsellors for women
to improve child health. BMC Pediatrics 2012;12:105. who want to breastfeed. Personal communication 2001.
[3414347] [3414172]
Ekstrom 2012 {published data only} Grossman 1988 {published data only}
∗
Ekstrom A, Kylberg E, Nissen E, Ekstrom A, Kylberg E, Grossman LK, Harter C, Kay A. Prenatal interventions
Nissen E. A process-oriented breastfeeding training program increase breast-feeding among low-income women.
for healthcare professionals to promote breastfeeding: American Journal of Diseases of Children 1988;142:404.
an intervention study: A process-oriented breastfeeding [3414174]
training program for healthcare professionals to promote Gurneesh 2009 {published data only}
breastfeeding: an intervention study. Breastfeeding Medicine Gurneesh S, Ellora D. Effect of antenatal expression of
2012;7(2):85–92. [3414160] breast milk at term to improve lactational performance: a
Ekstrom AC, Thorstensson S. Nurses and midwives prospective study. Journal of Obstetrics and Gynecology of
professional support increases with improved attitudes - India 2009;59(4):308–11. [3414176]
design and effects of a longitudinal randomized controlled
process-oriented intervention. BMC Pregnancy and Haider 2000 {published data only}
Childbirth 2015;15(1):275. [4446966] Haider R, Ashworth A, Kabir I, Huttly SRA. Effect
of community-based peer counsellors on exclusive
Feldman 1987 {published data only} breastfeeding practices in Dhaka, Bangladesh: a randomised
Feldman E, Hurst M. Outcomes and procedures in low risk controlled trial. Lancet 2000;356(9242):1643–7.
birth: a comparison of hospital and birth center settings. [3414178]
Birth 1987;14:18–24. [3414162]
Hanafi 2014 {published data only}
Forster 2011 {published data only} Hanafi MI, Shalaby SAH, Falatah N, El-Ammari H. Impact
Forster DA, McEgan K, Ford R, Moorhead A, Opie G, of health education on knowledge of, attitude to and
Walker S, et al. Diabetes and antenatal milk expressing: a practice of breastfeeding among women attending primary
pilot project to inform the development of a randomised health care centres in Almadinah Almunawwarah, kingdom
controlled trial. Midwifery 2011;27(2):209–14. [3414164] of Saudi Arabia: controlled pre-post study. Journal of Taibah
Froozani 1999 {published data only} University Medical Sciences 2014;9(3):187–93. [3414373]
Froozani MD, Permehzadeh K, Motlagh AR, Golestan B. Harvey 1996 {published data only}
Effect of breastfeeding education on the feeding pattern Harvey S, Jarrell J, Brant R, Stainton C, Rach D. A
and health of infants in their first 4 months in the Islamic randomised controlled trial of nurse-midwifery. Birth 1996;
Republic of Iran. Bulletin of the World Health Organization 23:128–35. [3414180]
1999;77(5):381–5. [3414166] Hegedus 2000 {published data only}
Garcia-Montrone 1996 {published data only} Hegedus Jungvirth M, Krcmar N, Smec D. Results of
Garcia-Montrone V, de Rose JC. An education experience breastfeeding promotion in the county of Medimurje.
for promoting breast-feeding and infant stimulation by low- Paediatria Croatica 2000;44(3):91–4. [3414182]
Henderson 2001 {published data only} der Geburt und postnatal]. Gynakologische Praxis 2005;29
Henderson A, Stamp G, Pincombe J. Postpartum (1):109–14. [3414192]
positioning and attachment education for increasing Kastner R, Gingelmaier A, Langer B, Grubert TA, Hartl K,
breastfeeding: a randomized trial. Birth 2001;28(4): Stauber M. Mother-child relationship before, during and
236–42. [3414184] after birth [Die Mutter–Kind–Beziehung pranatal, unter
Hirschhorn 2015 {published data only} der Geburt und postnatal]. Padiatrische Praxis 2005;67(1):
Hirschhorn LR, Semrau K, Kodkany B, Churchill R, 13–8. [3414193]
Kapoor A, Spector J, et al. Learning before leaping: Kistin 1990 {published data only}
integration of an adaptive study design process prior to Kistin N, Benton D, Rao S, Sullivan M. Breast-feeding rates
initiation of BetterBirth, a large-scale randomized controlled among black urban low-income women: effect of prenatal
trial in Uttar Pradesh, India. Implementation Science 2015; education. Pediatrics 1990;86(5):741–6. [3414195]
10(1):117. [3414375]
Kojuri 2009 {published data only}
Hives-Wood 2013 {published data only} Kojuri MD, Sakakky M, Hosseini F, Kherkhah M.
Hives-Wood S. Trial will test whether shopping vouchers Comparison of the effect of two methods of home visit
encourage breast feeding. BMJ (Clinical research ed.) 2013; for the promotion of exclusive breastfeeding in caesarean
347:F6807. [3414377] section mothers in Iran university of medical sciences 2008.
Hopkinson 2009 {published data only} International Journal of Gynecology & Obstetrics 2009;107
Hopkinson J, Konefal Gallagher M. Assignment to (Suppl 2):S150. [3414197]
a hospital-based breastfeeding clinic and exclusive
Kools 2005 {published data only}
breastfeeding among immigrant Hispanic mothers: a
Kools EJ, Thijs C, Kester ADM, van den Brandt PA, de
randomized, controlled trial. Journal of Human Lactation
Vries H. A breast-feeding promotion and support program
2009;25(3):287–96. [3414186]
a randomized trial in the Netherlands. Preventive Medicine
Howard 2000 {published data only} 2005;40:60–70. [3414199]
Howard C, Howard F, Lawrence R, Andresen E, DeBlieck
Kramer 2001 {published data only}
E, Weitzman M. Office prenatal formula advertising and its ∗
Kramer MS, Chalmers B, Hodnett ED, Sevkovskaya Z,
effect on breast-feeding patterns. Obstetrics & Gynecology
Kzikovich I, Shapiro S, et al. Promotion of breastfeeding
2000;95(2):296–303. [3414092]
intervention trial (PROBIT): a randomized trial in
Ijumba 2015 {published data only}
the Republic of Belarus. JAMA 2001;285(4):413–20.
Ijumba P, Doherty T, Jackson D, Tomlinson M, Sanders D,
[3414201]
Swanevelder S, et al. Effect of an integrated community-
Kramer MS, Matush L, Vanilovich I, Platt RW, Bogdanovich
based package for maternal and newborn care on feeding
N, Sevkovskaya Z, et al. A randomized breast-feeding
patterns during the first 12 weeks of life: a cluster-
promotion intervention did not reduce child obesity in
randomized trial in a South African township. Public Health
Belarus. Journal of Nutrition 2009;139(2):417S–21S.
Nutrition 2015;18(14):2660–8. [3414381]
[3414202]
Jahan 2014 {published data only}
Labarere 2011 {published data only}
Jahan K, Roy SK, Israt S, Ferdouse K, Salam SB. Impact of
Labarere J, Gelbert-Baudino N, Laborde L, Arragain D,
nutrition education on pregnancy weight gain and birth
Schelstraete C, Francois P. CD-ROM-based program for
outcome. Annals of Nutrition & Metabolism 2013;63(Suppl
breastfeeding mothers. Maternal & Child Nutrition 2011;7
1):756, Abstract no: PO1043. [4446967]
∗ (3):263–72. [3414395]
Jahan K, Roy SK, Mihrshahi S, Sultana N, Khatoon S,
Roy H, et al. Short-term nutrition education reduces low Lakin 2015 {published data only}
birthweight and improves pregnancy outcomes among Lakin A, Sutter MB, Magee S. Newborn well-child visits
urban poor women in Bangladesh. Food and Nutrition in the home setting: a pilot study in a family medicine
Bulletin 2014;35(4):414–21. [3414387] residency. Family Medicine 2015;47(3):217–21. [3414397]
Junior 2007 {published data only} Langer 1996 {published data only}
Junior WS, Martinez FE. Effect of intervention on the rates Langer A, Farnot U, Garcia C, Barros F, Victora C, Belizan
of breastfeeding of very low birth weight newborns. Jornal JM, et al. The Latin American trial of psychosocial support
de Pediatria 2007;83(6):541–6. [3414188] during pregnancy: effects on mother’s wellbeing and
Kaplowitz 1983 {published data only} satisfaction. The Latin American Network for Perinatal
Kaplowitz DD, Olson CM. The effect of an education and Reproductive Research (LANPER). Social Science and
program on the decision to breastfeed. Journal of Nutrition Medicine 1996;42(11):1589–97. [3414204]
Education 1983;15(2):61–5. [3414190] Langer 1998 {published data only}
Kastner 2005 {published data only} Langer A, Campero L, Garcia C, Reynoso S. Effects
Kastner R, Gingelmaier A, Langer B, Grubert TA, Hartl K, of psychosocial support during labour and childbirth
Stauber M. Mother-child relationship before, during and on breastfeeding, medical interventions, and mothers’
after birth [Die Mutter–Kind–Beziehung pranatal, unter wellbeing in a Mexican public hospital: a randomised
clinical trial. British Journal of Obstetrics and Gynaecology Matilla Mont 1999 {published data only}
1998;105(10):1056–63. [3414206] Matilla Mont M, Rios Jimenez A. Nursing and maternal
Lavender 2005 {published data only} breast feeding. Enfermeria Clinica 1999;9(3):93–7.
Lavender T. Breastfeeding: expectations versus reality. 10th [3414223]
International Conference of Maternity Care Researchers; Mattar 2007 {published data only}
2004 June 13-16; Lund, Sweden. 2004:12. [3414209] Mattar CN, Chong YS, Chan YS, Chew A, Tan P, Chan YH,
∗
Lavender T, Baker L, Smyth R, Collins S, Spofforth A, et al. Simple antenatal preparation to improve breastfeeding
Dey P. Breastfeeding expectations versus reality: a cluster practice: a randomized controlled trial. Obstetrics &
randomised controlled trial. BJOG: an international journal Gynecology 2007;109(1):73–80. [3414225]
of obstetrics and gynaecology 2005;112:1047–53. [3414208]
Loh 1997 {published data only} Maycock 2013 {published data only}
∗
Loh NR, Kelleher CC, Long S, Loftus BG. Can we increase Maycock B, Binns CW, Dhaliwal S, Tohotoa J, Hauck Y,
breast feeding rates?. Irish Medical Journal 1997;90(3): Burns S, et al. Education and support for fathers improves
100–1. [3414213] breastfeeding rates: a randomized controlled trial. Journal of
Human Lactation 2013;29(4):484–90. [3414408]
Lucchini 2013 {published data only}
Tohotoa J, Maycock B, Hauck YL, Dhaliwal S, Howat
Lucchini C, Uribe TC, Villarroel PL, Rojas RA. Randomized
P, Burns S. Can father inclusive practice reduce paternal
controlled clinical trial evaluating determinants of successful
postnatal anxiety? A repeated measures cohort study using
breastfeeding: Follow-up two months after comprehensive
the hospital anxiety and depression scale. BMC Pregnancy
intervention versus standard care delivery [Determinantes
and Childbirth 2012;12:75. [4446968]
para una lactancia materna exitosa: Intervencion integral
vs cuidado estandar. Ensayo clinico aleatorio controlado]. McEnery 1986 {published data only}
Revista Chilena de Pediatria 2013;84(2):138–44. [3414401] McEnery G, Rao KPS. The effectiveness of antenatal
MacVicar 1993 {published data only} education of Pakistani and Indian women living in this
MacVicar J, Dobbie G, Owen-Johnstone L, Jagger C, country. Child: Care, Health and Development 1986;12:
Hopkins M, Kennedy J. Simulated home delivery in 385–99. [3414227]
hospital: a randomised controlled trial. British Journal of
McInnes 2000 {published data only}
Obstetrics and Gynaecology 1993;98:316–23. [3414215]
McInnes RJ, Love JG, Stone DH. Evaluation of a
Mahmood 2011 {published data only} community-based intervention to increase breastfeeding
Mahmood I, Jamal M, Khan N. Effect of mother-infant prevalence. Journal of Public Health Medicine 2000;22(2):
early skin-to-skin contact on breastfeeding status: A 138–45. [3414229]
randomized controlled trial. Journal of the College of
Physicians and Surgeons Pakistan 2011;21(10):601–5. McLachlan 2016 {published data only}
[3414217] ACTRN12607000073404. A randomised trial comparing
One-to-One midwifery care with standard hospital
Martens 2000 {published data only}
maternity care for women at low risk, in order to decrease
Martens PJ. Does breastfeeding education affect nursing
operative birth and other interventions and increase the
staff beliefs, exclusive breastfeeding rates and Baby Friendly
duration of breastfeeding and women’s satisfaction with
Hospital Initiative compliance? The experience of a small
care, with no increase in costs of care. anzctr.org.au/Trial/
rural Canadian hospital. Journal of Human Lactation 2000;
Registration/TrialReview.aspx?id=81762 (first received 15
16:309–18. [3414219]
December 2006). [3414416]
Martens 2001 {published data only} Davey M, McLachlan H, Forster D. Timing of admission
Martens PJ. The effect of breastfeeding education on and selected aspects of intrapartum care: Relationship with
adolescent beliefs and attitudes: a randomized school caesarean section in the COSMOS (Caseload Midwifery)
intervention in the Canadian Ojibwa community of trial. Women & Birth 2013;26(Suppl 1):S3. [3414410]
Sagkeeng. Journal of Human Lactation 2001;17(3):245–55. Davey MA, McLachlan L, Forster D, Flood M. Influence of
[3414221] timing of admission in labour and management of labour
Martin 2013 {published data only} on method of birth: results from a randomised controlled
∗
Martin RM, Patel R, Kramer MS, Guthrie L, Vilchuck trial of caseload midwifery (COSMOS trial). Midwifery
K, Bogdanovich N, et al. Effects of promoting longer-term 2013;29(12):1297–302. [3414411]
and exclusive breastfeeding on adiposity and insulin-like Flood M, Forster DA, Davey MA, McLachlan HL. Serious
growth factor-I at age 11.5 years: a randomized trial. JAMA adverse event monitoring in a rct of caseload midwifery
2013;309(10):1005–13. [3414405] (cosmos). Journal of Paediatrics and Child Health 2012;48
Martin RM, Patel R, Kramer MS, Vilchuck K, Bogdanovich (Suppl 1):113. [3414412]
N, Sergeichick N, et al. Effects of promoting longer-term McLachlan H, Forster D, Davey MA. The effect of caseload
and exclusive breastfeeding on cardiometabolic risk factors midwifery on women’s experience of labour and birth:
at age 11.5 years: a cluster-randomized, controlled trial. results from the COSMOS randomised controlled trial.
Circulation 2014;129(3):321–9. [3414406] International Confederation of Midwives 30th Triennial
Congress. Midwives: Improving Women’s Health; 2014 Morhason-Bello 2009 {published data only}
June 1-4; Prague, Czech Republic. 2014:C085. [3414415] Morhason-Bello IO, Adedokun BO, Ojengbede OA. Social
McLachlan H, Forster D, Davey MA, Farrell T, Gold L, support during childbirth as a catalyst for early breastfeeding
Oats J, et al. A randomised controlled trial of caseload initiation for first-time Nigerian mothers. International
midwifery for women at low risk of medical complications Breastfeeding Journal 2009;4:16. [3414238]
∗
(COSMOS) - primary and secondary outcomes. Women Morhason-Bello IO, Adedokun BO, Ojengbede OA,
and Birth 2011;24 Suppl 1:S13. [3414414] Olayemi O, Oladokun A, Fabamwo AO. Assessment of
McLachlan H, Forster D, Davey MA, Gold L, Biro MA, the effect of psychosocial support during childbirth in
Flood M, et al. The effect of caseload midwifery on women’s Ibadan, south-west Nigeria: a randomised controlled trial.
experience of labour and birth: results from the COSMOS Australian and New Zealand Journal of Obstetrics and
randomised controlled trial. Women & Birth 2013;26(Suppl Gynaecology 2009;49(2):145–50. [3414237]
1):S13. [3414413]
∗
Morrow 1999 {published data only}
McLachlan HL, Forster DA, Davey MA, Farrell T, Flood
Morrow AL, Guerrero ML, Shults J, Calva JJ, Lutter C,
M, Shafiei T, et al. The effect of primary midwife-led
Bravo J, et al. Efficacy of home-based peer counselling to
care on women’s experience of childbirth: results from
promote exclusive breastfeeding: a randomized controlled
the COSMOS randomised controlled trial. BJOG: an
trial. Lancet 1999;353(9160):1226–31. [3414240]
International Journal of Obstetrics and Gynaecology 2016;
123:465–74. [3414417] Nasehi 2012 {published data only}
McLachlan HL, Forster DA, Davey MA, Farrell T, Gold L, Nasehi MM, Farhadi R, Ghaffari V, Ghaffari-Charati M.
Biro MA, et al. Effects of continuity of care by a primary The effect of early breastfeeding after cesarean section on
midwife (caseload midwifery) on caesarean section rates in the success of exclusive breastfeeding. HealthMED 2012;6
women of low obstetric risk: the COSMOS randomised (11):3597–601. [3414423]
controlled trial. BJOG: an international journal of obstetrics NCT00393640 {published data only}
and gynaecology 2012;119(12):1483–92. [3414418] NCT00393640. Early and regular breast milk expression
McLachlan HL, Forster DA, Davey MA, Farrell T, Gold to help establish lactation after delivery: a randomized
L, Oats J, et al. A randomised controlled trial of caseload controlled trial. clinicaltrials.gov/show/NCT00393640
midwifery for women at low risk of medical complications (first received 27 October 2006). [3414154]
(COSMOS): maternal and infant outcomes. Journal
NCT01916603 {published data only}
of Paediatrics and Child Health 2011;47(Suppl 1):33.
NCT01916603. Diet, physical activity and breastfeeding
[3414419]
intervention on maternal nutrition, offspring growth and
McLachlan HL, Forster DA, Davey MA, Farrell T, Gold
development. clinicaltrials.gov/ct2/show/NCT01916603
L, Waldenstrom U, et al. A randomised controlled trial
(first received 4 July 20130. [3414451]
of caseload midwifery for women at low risk of medical
complications (COSMOS): women’s satisfaction with care. NCT02162498 {published data only}
Journal of Paediatrics and Child Health 2012;48(Suppl 1): NCT02162498. Effect of feeding buddies on adherence
41–2. [3414420] to WHO PMTCT guidelines in South Africa [Study to
McLachlan HL, Forster DA, Davey MA, Lumley J, Farrell determine the impact of feeding buddies on adherence
T, Oats J, et al. COSMOS: comparing standard maternity to WHO PMTCT guidelines in South Africa].
care with one-to-one midwifery support: a randomised clinicaltrials.gov/ct2/show/NCT02162498 (first received
controlled trial. BMC Pregnancy and Childbirth 2008;8:35. 10 June 2014). [3414383]
[3414421] Nguyen 2014 {published data only}
McQueen 2011 {published data only} Nguyen PH, Menon P, Keithly SC, Kim SS, Hajeebhoy
McQueen KA, Dennis CL, Stremler R, Norman CD. A N, Tran LM, et al. Program impact pathway analysis of a
pilot randomized controlled trial of a breastfeeding self- social franchise model shows potential to improve infant
efficacy intervention with primiparous mothers. JOGNN: and young child feeding practices in Vietnam. Journal of
Journal of Obstetric, Gynecologic and Neonatal Nursing 2011; Nutrition 2014;144(10):1627–36. [3414425]
40:35–46. [3414231] Nikodem 1998 {published data only}
Moran 2000 {published data only} Nikodem VC, Nolte AGW, Wolman W, Gulmezoglu AM,
Moran VH, Bramwell R, Dykes F, Dinwoodie K. An Hofmeyr GJ. Companionship by a lay labour supporter to
evaluation of skills acquisition on the WHO/UNICEF modify the clinical birth environment: long-term effects
Breastfeeding Management Course using the pre-validated on mother and child. Curationis: South African Journal of
Breastfeeding Support Skills Tool (BeSST). Midwifery 2000; Nursing 1998;21(1):8–12. [3414242]
16(3):197–203. [3414233] Noel-Weiss 2006 {published data only}
Moreno-Manzanares 1997 {published data only} Noel-Weiss J, Rupp A, Cragg B, Bassett V, Woodend AK.
Moreno-Manzanares L, Cabrera-Sanz MT, Garcia-Lopez Randomized controlled trial to determine effects of prenatal
L. Breast feeding [Lactancia materna]. Revista Rol de breastfeeding workshop on maternal breastfeeding self-
Enfermeria 1997;20(227-228):79–84. [3414235] efficacy and breastfeeding duration. Journal of Obstetric,
Gynecologic and Neonatal Nursing 2006;35(5):616–24. income mothers. Pediatric Academic Societies Annual
[3414244] Meeting; 2007 May 5-8; Toronto, Canada 2007. [3414263]
Nor 2012 {published data only} Rea 1999 {published and unpublished data}
Nor B, Ahlberg BM, Doherty T, Zembe Y, Jackson D, Rea MF, Venancio SI, Marines JC, Savage F. Counselling on
Ekstrom EC, et al. Mother’s perceptions and experiences of breastfeeding: assessing knowledge and skills. Bulletin of the
infant feeding within a community-based peer counselling World Health Organization 1999;77(6):492–8. [3414265]
intervention in South Africa. Maternal & Child Nutrition
Redman 1995 {published data only}
2012;8(4):448–58. [3414246] Redman S, Watkins J, Evans L, Lloyd D. Evaluation of
Oakley 1990 {published data only} an Australian intervention to encourage breastfeeding in
Oakley A, Rajan L, Grant AM. Social support and primiparous women. Health Promotion International 1995;
pregnancy outcome. British Journal of Obstetrics and 10(2):101–13. [3414267]
Gynaecology 1990;97:155–62. [3414248] Reifsnider 1996 {published and unpublished data}
Page 1999 {published data only} Reifsnider E, Eckhart D. Prenatal breastfeeding education:
Page L, McCourt C, Beake S, Vail A, Hewison J. Clinical its effect on breastfeeding among WIC participants. Journal
interventions and outcomes of one-to-one midwifery of Human Lactation 1997;13(2):121–5. [3414269]
practice. Journal of Public Health Medicine 1999;21(3): Ross 1983 {published data only}
243–8. [3414250] Ross SM, Loening WEK, Middelkoop AV. Breast-feeding -
Petrova 2009 {published data only} evaluation of a health education programme. South African
Petrova A, Ayers C, Stechna S, Gerling JA, Mehta R. Medical Journal 1983;64(3):361–2. [3414271]
Effectiveness of exclusive breastfeeding promotion in Rossiter 1994 {published data only}
low-income mothers: a randomized controlled study. Rossiter JC. The effect of a culture-specific education
Breastfeeding Medicine 2009;4(2):63–9. [3414252] program to promote breastfeeding among Vietnamese
Philipp 2004 {published data only} women in Sydney. International Journal of Nursing Studies
Merewood A, Phillipp BL, Chamberlain LB, Malone KL, 1994;31(4):369–79. [3414273]
Cook JT, Bauchner H. Using peer support to improve Schafer 1998 {published data only}
breastfeeding rates among premature infants: an RCT. Schafer E, Vogel MK, Viegas S, Hausafus C. Volunteer peer
Pediatric Academic Societies Annual Meeting; 2005 May counselors increase breastfeeding duration among rural low-
14-17; Washington DC, USA. 2005. [3414254] income women. Birth 1998;25(2):101–6. [3414275]
∗
Philipp BL, Merewood A, Malone KL, Chamberlain
Schlickau 2005 {published data only}
LB, Cook JT, Bauchner H. Effect of NICU-based peer
Schlickau J, Wilson M. Development and testing of a
counselors on breastfeeding duration among premature
prenatal breastfeeding education intervention for Hispanic
infants [abstract]. Pediatric Research 2004;55 Suppl:73.
women. Journal of Perinatal Education 2005;14(4):24–35.
[3414255]
[3414277]
Pisacane 2005 {published data only} ∗
Schlickau JM. Prenatal Breastfeeding Education: an
Pisacane A, Continisio GI, Aldinucci M, D’amora S, Intervention for Pregnant Immigrant Hispanic Women [thesis].
Continisio P. A controlled trial of the father’s role in Nebraska: University of Nebraska, 2005. [4446969]
breastfeeding promotion. Pediatrics 2005;116:e494–8.
Schwartz 2015 {published data only}
[3414257]
Schwartz R, Vigo A, de Oliveira LD, Giugliani ERJ. The
Pobocik 2000 {published data only} effect of a pro-breastfeeding and healthy complementary
Pobocik RS, Benavente JC, Schwab AC, Boudreau N, feeding intervention targeting adolescent mothers and
Morris CH, Houston MS. Effect of a breastfeeding grandmothers on growth and prevalence of overweight
education and support program on breastfeeding initiation of preschool children. PLOS One 2015;10(7):e0131884.
and duration in a culturally diverse group of adolescents. [3414435]
Journal of Nutrition Education 2000;32(3):139–45. Schy 1996 {published data only}
[3414259] Schy DS, Maglaya CF, Mendelson SG, Race KE, Ludwig-
Prakhin 2001 {published data only} Beymer P. The effects of in-hospital lactation education on
Prakhin EI. Role of educational and scientific programs in breastfeeding practice. Journal of Human Lactation 1996;
realization of the concept of healthy nutrition in Siberia 12:117–22. [3414279]
[Rol’ obrazovatel’nykh i nauchnykh programm v realizatsii Sciacca 1995 {published and unpublished data}
kontseptsii zdorovogo pitaniia v Sibiri]. Vorprosy Pitaniia Sciacca JP, Dube DA, Phipps BL, Ratliff MI. A breast
2001;70(2):42–4. [3414261] feeding education and promotion program: effects on
Pugh 2007 {published data only} knowledge, attitudes and support for breast feeding. Journal
Pugh LC, Nanda JP, Frick KD, Sharps PW, Spatz DL, of Community Health 1995;20(6):473–90. [3414281]
∗
Serwint JR, et al. A randomized controlled community- Sciacca JP, Phipps BL, Dube CA, Ratliff MI. Influences on
based trial to improve breastfeeding among urban low- breast-feeding by lower-income women: an incentive-based,
partner-supported educational program. Journal of the Turan 2001 {published data only}
American Dietetic Association 1995;95:323–8. [3414282] Turan JM, Nalbant H, Bulut A, Sahip Y. Including
expectant fathers in antenatal education programmes in
Scott 1975 {published data only}
Istanbul, Turkey. Reproductive Health Matters 2001;9(18):
Scott Brown M, Hurlock JT. Preparation of the breast
114–25. [3414294]
for breast-feeding. Nursing Research 1975;24:448–51.
[3414284] Turnbull 1996 {published data only}
Turnbull D, Holmes A, Shields N, Cheyne H, Twaddle S,
Sellen 2012 {published data only}
Harper Gilmore W, et al. Randomised controlled trial of
Sellen DW, Kamau-Mbuthia E, Mbugua S, Webb Girard
efficacy of midwife-managed care. Lancet 1996;348:213–8.
AL, Lou W, Dennis CL, et al. Lessons learned in providing
[3414296]
peer support through cell phones and group meetings
to increase exclusive breastfeeding in Kenya. Sixteenth Tylleskar 2011 {published data only}
ISRHML Conference “Breastfeeding and the Use of Human Birungi N, Fadnes LT, Okullo I, Kasangaki A, Nankabirwa
Milk. Science and Practice”; 2012 September 27-October V, Ndeezi G, et al. Effect of breastfeeding promotion on
1; Trieste, Italy. 2012:Abstract no. A18. [3414437] early childhood caries and breastfeeding duration among 5
year old children in Eastern Uganda: A cluster randomized
Shaw 1999 {published data only}
trial. PLOS One 2015;10(5):e0125352. [4446971]
Shaw E, Kaczorowski J. The effect of a peer counseling
Chola L, Fadnes LT, Engebretsen IM, Tumwine JK,
program on breastfeeding initiation and longevity in a low-
Tylleskar T, Robberstad B, et al. Infant feeding survival and
income rural population. Journal of Human Lactation 1999;
Markov transition probabilities among children under age 6
15(1):19–25. [3414286]
months in Uganda. American Journal of Epidemiology 2013;
Sisk 2004 {published data only} 177(5):453–62. [4446972]
Sisk JE, Greer AL, Wojtowycz M, Pincus LB, Aubry RH. Chola L, Fadnes LT, Engebretsen IMS, Nkonki L,
Implementing evidence-based practice: evaluation of an Nankabirwa V, Sommerfelt H, et al. Cost-effectiveness of
opinion leader strategy to improve breast-feeding rates. peer counselling for the promotion of exclusive breastfeeding
American Journal of Obstetrics and Gynecology 2004;190: in Uganda. PLOS One 2015;10(11):e0142718. [4446973]
413–21. [3414288] Engebretsen I, Nankunda J, Nankabirwa V, Diallo A,
Fadnes L, Doherty T, et al. Early infant feeding practices in
Spinelli 2013 {published data only}
the Promise-EBF trial: promotion of exclusive breastfeeding
Spinelli MG, Endicott J, Goetz RR. Increased breastfeeding
by peer counsellors in three countries in Africa. Annals of
rates in black women after a treatment intervention.
Nutrition & Metabolism 2013;63(Suppl 1):709, Abstract
Breastfeeding Medicine 2013;8(6):479–84. [3414439]
no: PO940. [4446974]
Susin 2008 {published data only} Engebretsen IM, Jackson D, Fadnes LT, Nankabirwa V,
Susin LR, Giugliani ER. Inclusion of fathers in an Diallo AH, Doherty T, et al. Growth effects of exclusive
intervention to promote breastfeeding: impact on breastfeeding promotion by peer counsellors in sub-Saharan
breastfeeding rates. Journal of Human Lactation 2008;24 Africa: the cluster-randomised PROMISE EBF trial. BMC
(4):386–92. [3414290] Public Health 2014;14(1):633. [4446975]
Engebretsen IM, Jackson D, Fadnes LT, Nankabirwa V,
Talukder 2016 {published data only}
Diallo AH, Doherty T, et al. Is promotion of exclusive
Talukder S, Farhana D, Vitta B, Greiner T. In a rural
breastfeeding safe in sub-Sharan Africa with respect to child
area of Bangladesh, traditional birth attendant training
growth? Results from the cluster-randomised PROMISE
improved early infant feeding practices: a pragmatic cluster
EBF-trial. Sixteenth ISRHML Conference “Breastfeeding
randomized trial. Maternal & Child Nutrition 2016 [Epub
and the Use of Human Milk. Science and Practice”; 2012
ahead of print]. [3414447]
September 27-October 1; Trieste, Italy. 2012. [4446976]
Talukder SH, Greiner T, Dewey K, Haider R, Farhana
Engebretsen IMS, Nankabirwa V, Doherty T, Diallo
D, Chowdhury SS. Cost and effectiveness of training and
AH, Nankunda J, Fadnes LT, et al. Early infant feeding
supervision of frontline workers on early breastfeeding
practices in three African countries: the PROMISE-EBF
practices in Bangladesh. Sixteenth ISRHML Conference
trial promoting exclusive breastfeeding by peer counsellors.
“Breastfeeding and the Use of Human Milk. Science and
International Breastfeeding Journal 2014;9:19. [4446977]
Practice”; 2012 September 27-October 1; Trieste, Italy.
NCT00397150. PROMISE EBF: safety and efficacy of
2012:Abstract A106. [4446970]
exclusive breastfeeding promotion in the era of HIV in sub-
Toma 2001 {published data only} Saharan Africa. clinicaltrials.gov/ct2/show/NCT00397150
Toma TS, Monteiro CA. Assessment of the promotion of (first received 7 November 2006). [3414299]
∗
breastfeeding in public and private maternities of Sao Paulo Tylleskar T, Jackson D, Meda N, Engebretsen IM, Chopra
city, Brazil [Avaliacao da promocao do aleitamento materno M, Diallo AH, et al. Exclusive breastfeeding promotion by
nas maternidades publicas e privadas do Municipio de peer counsellors in sub-Saharan Africa (PROMISE-EBF): a
Sao Paulo]. Revista de Saude Publica 2001;35(5):409–14. cluster-randomised trial. Lancet 2011;378(9789):420–7.
[3414292] [3414298]
Vaidya 2005 {published data only} a randomized controlled trial of an educational intervention.
Vaidya K, Sharma A, Dhungel S. Effect of early mother-baby American Journal of Obstetrics and Gynecology 2004;191:
close contact over the duration of exclusive breastfeeding. 708–12. [3414120]
Nepal Medical College Journal: NMCJ 2005;7(2):138–40. Woolridge 1985 {published data only}
[3414301] Woolridge MW, Greasley V, Silpisornkosol S. The initiation
van den Bosch 1990 {published data only} of lactation: the effect of early versus delayed contact for
van den Bosch CA, Bullough CHW. Effect of early suckling suckling on milk intake in the first week post-partum. A
on term neonates’ core body temperature. Annals of Tropical study in Chiang Mai, Northern Thailand. Early Human
Paediatrics 1990;10:347–53. [3414303] Development 1985;12:269–78. [3414321]
Vianna 2011 {published data only}
Yotebieng 2015 {published data only}
Vianna MNS, Barbosa AP, Carvalhaes AS, Cunha AJLA.
Yotebieng M, Labbok M, Soeters HM, Chalachala
Music therapy may increase breastfeeding rates among
JL, Lapika B, Vitta BS, et al. Ten steps to successful
mothers of premature newborns: a randomized controlled
breastfeeding programme to promote early initiation and
trial [A musicoterapia pode aumentar os indices de
exclusive breastfeeding in DR Congo: a cluster-randomised
aleitamento materno entre maes de recem–nascidos
controlled trial. Lancet Global Health 2015;3(9):e546–55.
prematuros: um ensaio clinico randomizado controlado].
[3414459]
Jornal de Pediatria 2011;87(3):206–12. [3414305]
Zimmerman 1999 {published data only}
Volpe 2000 {published data only}
Zimmerman DR. You can make a difference: increasing
Volpe EM, Bear M. Enhancing breastfeeding initiation in
breastfeeding rates in an inner-city clinic. Journal of Human
adolescent mothers through the Breastfeeding Educated and
Lactation 1999;15(3):217–20. [3414323]
Supported Teen (BEST) Club. Journal of Human Lactation
2000;16(3):196–200. [3414307] References to studies awaiting assessment
Waldenstrom 1994 {published data only}
Waldenstrom U, Nilsson CA. No effect of birth centre care Bakhshi 2015 {published data only}
on either duration or experience of breast feeding but more Bakhshi M, Kordi M, Esmaeeli H. The effect of continuous
complications: findings from a randomised controlled trial. support during labor on the onset of lactogenesis stage II
Midwifery 1994;10:8–17. [3414309] in primiparas. Journal of Mazandaran University of Medical
Westphal 1995 {published data only} Sciences 2015;25(130):153–8. [3414331]
Taddei JA, Westphal MF, Venancio S, Bogus C, Souza S. Samieizadeh 2011 {published data only}
Breastfeeding training for health professionals and resultant Samieizadeh T, Sereshti M, Dashipur AR, Mohammadinia
changes in breastfeeding duration. Sao Paulo Medical N, Arzani A. The effect of supportive companionship on
Journal 2000;118(6):185–91. [3414311] Length of labor and desire to breastfeed in primiparous
Westphal MF, Taddei JAC, Venancio SI, Bogus CM. Breast- Women. Journal of Urmia Nursing & Midwifery Faculty
feeding training for health professionals and resultant 2011;9(4):1–9. [3414433]
institutional changes. Boletin De La Oficina Sanitaria
Panamericana 1996;120(4):304–15. [3414312] References to ongoing studies
∗
Westphal MF, Taddei JAC, Venancio SI, Bogus CM.
Breast-feeding training for health professionals and ISRCTN23019866 {published data only}
resultant institutional changes. Bulletin of the World Health ISRCTN23019866. Evaluating the family nurse
Organization 1995;73(4):461–8. [3414313] partnership programme in England: a randomised
Wiles 1984 {published data only} controlled trial. isrctn.com/ISRCTN23019866 (first
Wiles LS. The effect of prenatal breastfeeding education on received 24 March 2009). [3414461]
breastfeeding success and maternal perception of the infant. Kimani-Murage 2013 {published data only}
Journal of Obstetric, Gynecologic and Neonatal Nursing 1984; Kimani-Murage EW, Kyobutungi C, Ezeh AC, Wekesah F,
13(4):253–7. [3414315] Wanjohi M, Muriuki P, et al. Effectiveness of personalised,
Winterburn 2003 {published data only} home-based nutritional counselling on infant feeding
Winterburn S, Moyez J, Thompson J. Maternal practices, morbidity and nutritional outcomes among
grandmothers and support for breastfeeding. Journal of infants in Nairobi slums: study protocol for a cluster
Community Nursing 2003;17(12):4–9. [3414317] randomised controlled trial. Trials [Electronic Resource]
Winters 1973 {published data only} 2013;14:445. [3414391]
Winters N. The Relationship of Time of Initial Breast Feeding NCT02084680 {published data only}
to Success in Breast Feeding [MD thesis]. Washington: NCT02084680. A community trial to measure the effect
University of Washington, 1973. [3414319] of individual prenatal education with mobile phone
Wolfberg 2004 {published data only} consultations offered to pregnant women in Masindi and
Wolfberg AJ, Michels KB, Shields W, O’Campo P, Bronner Kiryandongo, Western Uganda. clinicaltrials.gov/ct2/show/
Y, Bienstock J. Dads as breastfeeding advocates: results from NCT02084680 (first received 10 March 2014). [3414403]
Williams 2015 {published data only} outcomes: a systematic review and meta-analysis. Acta
∗
Williams A, Chantry C, Dentz H, Kiprotich M, Null C, Paediatrica 2015;104:96–113.
Stewart C. Effectiveness of behavior change communication Copeland 2015
on maternal nutrition and breastfeeding practices within a Copeland L, Paranjothy S. Motivational interviewing: a
cluster randomized trial in rural Western Kenya. Journal of key approach for supporting the transition to parenthood
Human Lactation 2015;31(3):534–5. [3414457] including breastfeeding?. Perspective - NCT’s journal on
Williams AM, Chantry C, Dentz H, Kiprotich M, preparing parents for birth and early parenthood 2015;27:
Null C, Stewart CP. Effectiveness of behavior change 17–20.
communication on maternal nutrition and breastfeeding
Dyson 2006
practices within a cluster randomized trial in rural Western
Dyson L, Renfrew M, McFadden R, McCormick F, Herbert
Kenya. 17th Conference of the International Society for
G, Thomas J. Promotion of breastfeeding initiation and
Research in Human Milk and Lactation (ISRHML); 2014
duration. Evidence into Practice Briefing. London: NICE,
Oct 23-27; Kiawah Island, South Carolina, USA. 2014:
2006.
140. [4446978]
Dyson 2010
Additional references Dyson L, Renfrew MJ, McFadden A, McCormick F, Herbert
G, Thomas J. Policy and public health recommendations
Alderson 2004 to promote the initiation and duration of breast-feeding in
Alderson P, Green S, Higgins JPT, editor(s). Cochrane developed country settings. Public Health Nutrition 2010;
Reviewers’ Handbook 4.2.2 (updated March 2004). 13(1):137–44.
www.epidemiologia.anm.edu.ar/cochrane/pdf/
Fairbank 2000
handbook.pdf.
Fairbank L, O’Meara S, Renfrew MJ, Woolridge M, Sowden
Atkins 2004 AJ, Lister-Sharp D. A systematic review to evaluate the
Atkins D, Best D, Briss PA, Eccles M, Falck-Ytter Y, effectiveness of interventions to promote the initiation of
Flottorp S, et al. GRADE Working Group. Grading quality breastfeeding. Health Technology Assessment 2000; Vol. 4,
of evidence and strength of recommendations. BMJ 2004; issue 25:1–171.
328(7454):1490. Grassley 2007
Beake 2012 Grassley JS, Eschiti VS. Two generations learning together:
Beake S, Pellowe C, Dykes F, Schmied V, Bick D. A facilitating grandmothers’ support of breastfeeding.
systematic review of structured compared with non- International Journal of Childbirth Education 2007;22(3):
structured breastfeeding programmes to support the 23–6.
initiation and duration of exclusive and any breastfeeding in Heikkilä 2011
acute and primary health care settings. Maternal & Child Heikkilä K, Sacker A, Kelly Y, Renfrew MJ, Quigley MA.
Nutrition 2012;8:141–61. Breast feeding and child behaviour in the Millennium
Bolling 2007 Cohort Study. Archives of Disease in Childhood 2011;96:
Bolling K, Grant C, Hamlyn B, Thornton A. Infant 635.
Feeding Survey 2005. London: The Information Centre, Heikkilä 2014
Government Statistical Service, UK Health Departments, Heikkilä K, Kelly Y, Renfrew MJ, Sacker A, Quigley
2007. MA. Breastfeeding and educational achievement at age 5.
Bowatte 2015 Maternal & Child Nutrition 2014;10:92–101.
Bowatte G, Tham R, Allen K, Tan D, Lau M, Dai X, Hermann 2014
et al. Breastfeeding and childhood acute otitis media: a Herrmann K, Carroll K. An exclusively human milk diet
systematic review and meta-analysis. Acta Paediatrica 2015; reduces necrotizing enterocolitis. Breastfeeding Medicine
104:85–95. 2014;9(4):184–90.
Bryant 1990 Higgins 2011
Bryant C, Roy M. Best Start Training Program. Tampa: Best Higgins JP, Green S, editor(s). Cochrane Handbook
Start Inc, 1990. for Systematic Reviews of Interventions Version 5.1.0
CDCP 2010 (updated March 2011). The Cochrane Collaboration,
Centers for Disease Control and Prevention. Racial and 2011. Available from handbook.cochrane.org.
Ethnic Differences in Breastfeeding Initiation and Duration, Horta 2013
by State - National Immunization Survey, United States, Horta B, Victora C, World Health Organization. Short-
2004-2008. www.cdc.gov/mmwr/preview/mmwrhtml/ term effects of breastfeeding: a systematic review on the
mm5911a2.htm#tab1 (accessed 27 June 2016). benefits of breastfeeding on diarrhoea and pneumonia
Chowdhury 2015 mortality. www.who.int/maternal˙child˙adolescent/
Chowdhury R, Sinha B, Sankar MJ, Taneja S, Bhandari documents/breastfeeding˙short˙term˙effects/en/ (accessed
N, Rollins N, et al. Breastfeeding and maternal health prior to 27 October 2016).
Horta 2015a NEOVITA Study Group 2016
Horta BL, Loret de Mola C, Victora CG. Long-term NEOVITA Study Group. Timing of initiation, patterns of
consequences of breastfeeding on cholesterol, obesity, breastfeeding, and infant survival: prospective analysis of
systolic blood pressure and type 2 diabetes: a systematic pooled data from three randomised trials. Lancet Global
review and meta-analysis. Acta Paediatrica 2015;104:30–7. Health 2016;4(4):e266–75.
Horta 2015b Peres 2015
Horta BL, Loret de Mola C, Victora CG. Breastfeeding and Peres KG, Cascaes AM, Nascimento GG, Victora CG.
intelligence: a systematic review and meta-analysis. Acta Effect of breastfeeding on malocclusions: a systematic
Paediatrica 2015;104:14–9. review and meta-analysis. Acta Paediatrica 2015;104:
54–61.
Ingram 2004
Phipps 2006
Ingram J, Johnson D. A feasibility study of an intervention
Phipps B. Peer support for breastfeeding in the UK. British
to enhance family support for breast feeding in a deprived
Journal of General Practice 2006;56:166–7.
area in Bristol, UK. Midwifery 2004;20:367–79.
Piwoz 2015
Ip 2007
Piwoz EG, Huffman SL. The impact of marketing of breast-
Ip S, Chung M, Raman G, Chew P, Magula N, Devine
milk substitutes on WHO-recommended breastfeeding
D, et al. Breastfeeding and maternal and infant health
practices. Food and Nutrition Bulletin 2015;36(4):373–86.
outcomes in developed countries. Evidence Report/
Technology Assessment 2007;April(153):1–186. Pérez-Escamilla 2016
Pérez-Escamilla R, Martinez JL, Segura-Pérez S. Impact of
Jiang 2010 the baby-friendly hospital initiative on breastfeeding and
Jiang M, Foster EM, Gibson-Davis CM. The effect of WIC child health outcomes: a systematic review. Maternal &
on breastfeeding: a new look at an established relationship. Child Nutrition 2016;12:402–17.
Children and Youth Services Review 2010;32(2):264–73.
Quigley 2012
Jolly 2012 Quigley MA, Hockley C, Carson C, Kelly Y, Renfrew MJ,
Jolly K, Ingram L, Khan KS, Deeks JJ, Freemantle N, Sacker A. Breastfeeding is associated with improved child
MacArthur C. Systematic review of peer support for cognitive development: a population-based cohort study.
breastfeeding continuation: metaregression analysis of the Journal of Pediatrics 2012;160:25–32.
effect of setting, intensity, and timing. BMJ 2012;344:
Renfrew 2012a
d8287.
Renfrew MJ, Pokhrel S, Quigley M, McCormick F, Fox-
Lawton 2012 Rushby J, Dodds R, et al. Preventing Disease and Saving
Lawton R, Ashley L, Dawson S, Waiblinger D, Conner M. Resources: the potential contribution of increasing breastfeeding
Employing an extended theory of planned behaviour to rates in the UK. London: UNICEF, 2012.
predict breastfeeding intention, initiation, and maintenance
Renfrew 2012b
in White British and South-Asian mothers living in
Renfrew MJ, McCormick FM, Wade A, Quinn B, Dowswell
Bradford. British Journal of Health Psychology 2012;17:
T. Support for healthy breastfeeding mothers with healthy
854–71.
term babies. Cochrane Database of Systematic Reviews 2012,
Lutter 2013 Issue 5. [DOI: 10.1002/14651858.CD001141.pub4]
Lutter CK. The International Code of Marketing of Breast-
RevMan 2014 [Computer program]
milk Substitutes: lessons learned and implications for the
Nordic Cochrane Centre, The Cochrane Collaboration.
regulation of marketing of foods and beverages to children.
Review Manager 5 (RevMan 5). Version 5.2. Copenhagen:
Public Health Nutrition 2013;16:1879–84.
Nordic Cochrane Centre, The Cochrane Collaboration,
McAndrew 2012 2014.
McAndrew F, Thompson J, Fellows L, Large A, Speed Rollins 2016
M, Renfrew MJ. Infant Feeding Survey 2010. Leeds, UK: Rollins N, Bhandari N, Hajeebhoy N, Horton S, Lutter
Health and Social Care Information Centre, 2012. C, Martines J, et al. Why invest, and what it will take
McFadden 2016 to improve breastfeeding practices?. Lancet 2016;387:
McFadden A, Mason F, Baker J, Begin F, Dykes F, 491–504.
Grummer-Strawn L, et al. Spotlight on infant formula: Sankar 2015
coordinated global action needed. Lancet 2016;387 Sankar MJ, Sinha B, Chowdhury R, Bhandari N, Taneja S,
(10017):413–5. Martines J, et al. Optimal breastfeeding practices and infant
Moore 2012 and child mortality: a systematic review and meta-analysis.
Moore ER, Anderson GC, Bergman N, Dowswell T. Early Acta Pædiatrica 2015;104:3–13.
skin-to-skin contact for mothers and their healthy newborn Sinha 2015
infants. Cochrane Database of Systematic Reviews 2012, Issue Sinha B, Chowdhury R, Sankar MJ, Martines J, Taneja S,
5. [DOI: 10.1002/14651858.CD003519.pub3] Mazumder S, et al. Interventions to improve breastfeeding
outcomes: a systematic review and meta-analysis. Acta 2003.
Paediatrica 2015;104:114–34. WHO Data Bank 1996
Tham 2015 WHO Global Data Bank on Breast-Feeding. Breast-Feeding:
Tham R, Bowatte G, Dharmage S, Tan D, Lau M, Dai X, et The Best Start in Life. Geneva: World Health Organization,
al. Breastfeeding and the risk of dental caries: a systematic 1996.
review and meta-analysis. Acta Paediatrica 2015;104:
WHO/UNICEF 1989
62–84.
WHO/UNICEF. Protecting, Promoting and Supporting
UNICEF 2014 Breastfeeding: The Special Role of Maternity Services. Geneva:
UNICEF. Infant and Young Child Feeding. Current Status World Health Organization, 1989.
+ Progress. data.unicef.org/nutrition/iycf (accessed 27 June
2016). References to other published versions of this review
Victora 2016b
Victora CG, Bahl R, Barros AJ, França GV, Horton S, Dyson 2005
Krasevec J. Supplementary appendix to Victora CG, Bahl Dyson L, McCormick FM, Renfrew MJ. Interventions
R, Barros AJD, et al, for The Lancet Breastfeeding Series for promoting the initiation of breastfeeding. Cochrane
Group. Breastfeeding in the 21st century: epidemiology, Database of Systematic Reviews 2005, Issue 2. [DOI:
mechanisms, and lifelong effect. Lancet 2016;387:475–90. 10.1002/14651858.CD001688.pub2]
WHO 1981 Fairbank 1999
World Health Organization. International Code of Fairbank L, Lister-Sharpe D, Renfrew MJ, Woolridge MW,
Marketing of Breast Milk Substitutes. Geneva: World Health Sowden AJS, O’Meara S. Interventions for promoting
Organization, 1981. the initiation of breastfeeding. Cochrane Database of
WHO 2003 Systematic Reviews 1999, Issue 3. [DOI: 10.1002/
World Health Organization. Global Strategy for Infant and 14651858.CD001688]
∗
Young Child Feeding. Geneva: World Health Organization, Indicates the major publication for the study

CHARACTERISTICS OF STUDIES
Characteristics of included studies [ordered by study ID]
Brent 1995
Methods Randomisation by permuted block.

Analysis was by intention-to-treat.
Participants 108 English-speaking, nulliparous, pregnant women attending a prenatal clinic, regard-
less of infant-feeding preference were recruited into study. Participants stratified by age
into 3 groups (less than 20, 20-29, or at least 30 years)
Interventions Experimental group: (N = 51).

Breastfeeding education and support provided throughout the prenatal and postpartum
periods and into the first year of the child’s life. Education consisted of 2 to 4 individual
10 to 15-minute sessions with a lactation consultant discussing the benefits and practice
of breastfeeding. Content of sessions was based on the women’s needs and interests.
After delivery, mothers were followed up with daily inpatient rounds by the lactation
consultant. Further follow-up consisted of a telephone call 48 hours after discharge, a
visit to the lactation clinic at 1 week and lactation consultation present at each health
supervision visit until weaning or when the infant was 1 year of age, whichever came
first.
Professional education was directed at nursing and medical staff who interacted with the
breastfeeding dyad
Control group: (N = 57).
Routine care, consisting of optional prenatal breastfeeding classes; postpartum breast-
feeding instruction by nurses and doctors; outpatient follow-up in the paediatric ambu-
latory department
Outcomes Incidence of breastfeeding in hospital.

Incidence of breastfeeding at 2 weeks.
Incidence of breastfeeding at 2 months.
Incidence of breastfeeding at 6 months.
Median duration of breastfeeding.
Subgroup analysis for women who indicated at the first prenatal visit that they planned
to formula feed or were undecided
Notes To determine if a comprehensive breastfeeding promotion programme increased the

incidence and duration of breastfeeding in a low-income population
Risk of bias
Bias Authors’ judgement Support for judgement
Random sequence generation (selection Low risk Patients, stratified by age were randomised
bias) into the intervention and control groups by
using a blocked randomisation procedure
….randomisation was performed in block
sizes of 8, pg 799

Brent 1995 (Continued)
Allocation concealment (selection bias) Unclear risk Unclear whether allocation concealment
was adequate. Allocation of participants to
either intervention or control groups was
not clearly described. According to the au-
thors, “patients were randomised into the
intervention and control groups by using
a blocked randomisation procedure”. They
say further that “patients assigned to the in-
tervention group required a minimum of
two prenatal lactation consultations to be
included in the sample”. It is unclear if this
criteria was an overall eligibility criteria for
the study or if it was applied to the inter-
vention group, pg 799
Blinding of participants and personnel High risk Inadequate due to non-blinded study. In-
(performance bias) tervention was conducted by lactation con-
All outcomes sultant who also administered the question-
naires
Blinding of outcome assessment (detection High risk Outcome assessors were not blinded to
bias) group allocations. “Data were collected by
All outcomes questionnaire that were administered in
person, not blinded by the lactation con-
sultant at the first prenatal visit…”. Out-
come data were reported by mothers and it
is possible that reports may have been bi-
ased
Incomplete outcome data (attrition bias) Low risk Adequate, breastfeeding initiation reported
All outcomes for all 108 women in the study. Although
the study tables could not be read easily
because it was darkened during production,
the participant numbers seem balanced and
reasons were provided for exclusions made
from the intervention group
Selective reporting (reporting bias) Unclear risk Study protocol was not available to assess
the prespecified outcomes
Other bias High risk Mothers in the intervention group were

found to have had an increased rate of com-
plications of pregnancy compared to the
control group. This may reflect some fun-
damental differences in the characteristics
of the women in both groups, pg 780

Caulfield 1998
Methods Method of allocation of the 4 clinics: 4 slips of paper labelled with 1 of 4 clinics randomly
selected from pot for centralised allocation to a pre-ordered list of comparison groups
Analysis was not by intention-to-treat.
Participants 4 clinics administered through the Johns Hopkins University WIC programme, that had
similar rates for ethnicity (90.4% to 96.1% African-American) and breastfeeding rates
at 1 month (2.0% to 5.9% in 1991)
Women were recruited between April 1992 and January 1994 as they registered for
prenatal care at the 4 clinics. 674 women were eligible. 242 had complete data (36%)
and only these were included in the results. Differences were noted by clinic in parity,
education and employment status before and during pregnancy of the included women
Interventions 2 x 2 factorial design.

Control (N = 57). Routine WIC services and nutrition education.
Video intervention (N = 64). Breastfeeding motivational video, based on Best Start
video, consisted of 8 trigger vignettes 2 to 5 minutes in length, about benefits of and
major benefits to breastfeeding, played continuously in the waiting area without staff
supervision. Discussion with service provider following video was encouraged. Posters
displayed in clinic areas and relevant sites
Peer counselling intervention (N = 55). Women interested in breastfeeding received
personalised information and support on breastfeeding issues of concern specific to each
participant. Women received counselling 3 times during pregnancy. WIC counsellors
were former WIC clients, had successfully breastfed and completed 5-week training
programme.
Video and peer counsellor (N = 66). All the components described above
Outcomes Breastfeeding initiation.

Breastfeeding initiation by infant-feeding intention at enrolment
Breastfeeding at 7 to 10 days for those who initiated.
Notes Not included in the meta-analysis on statistician’s advice, because with only 1 clinic in
each group, it is not possible to calculate the standard error of difference
Risk of bias
Random sequence generation (selection Unclear risk Method of sequence generation not de-
bias) scribed. Information limited to “four clin-
ics were randomly assigned to control and
treatment groups.”
Allocation concealment (selection bias) Unclear risk Allocation concealment procedures not de-
scribed. Information limited to “four clin-
ics were randomly assigned to control and
treatment groups.”

Caulfield 1998 (Continued)
Blinding of participants and personnel High risk It is not stated whether mothers and per-
(performance bias) sonnel were blinded. However, given the
All outcomes nature of the intervention it would not have
been possible to blind staff
Blinding of outcome assessment (detection Unclear risk Data were collected by trained interviewers
bias) but it is not stated whether the interviewers
All outcomes were blinded
Incomplete outcome data (attrition bias) High risk 548 women were enrolled in the study and
All outcomes 273 remained in the study to the end, so
50.2% were lost to follow-up
Selective reporting (reporting bias) Unclear risk No protocol or evidence of predefined out-
comes to judge this domain
Other bias High risk Baseline differences in parity, education,

and employment status before and dur-
ing pregnancy, between women enrolled at
each clinic
Chapman 2004
Methods Recruiter not the same as peer counsellors. Computerised random allocation of weekly
cases: appears to be on-site but not stated. Data entry of cases likely to be Research
Assistant who recruited but not likely to know how case would be allocated as SPSS
random selection
Data collection on infant-feeding practices, sources of breastfeeding support and demo-
graphics via face-to-face or telephone interviews by a researcher during the hospital stay
or monthly calls thereafter
Participants Pregnant women attending Hartford Hospital, Connecticut, USA, on 1 of 3 days a week
when recruitment conducted between July 2000 and August 2002.
Inclusion (prenatal) over 18 years old, considering breastfeeding, low-income.
Inclusion (postpartum) healthy full-term singleton infant, no maternal history of HIV.
Exclusion: infants admitted to special care.
Interventions Control group (N = 75).

Routine prenatal breastfeeding education consisted of individualised breastfeeding in-
formation offered in response to women’s questions, and written breastfeeding materials
from the prenatal clinic. Routine perinatal breastfeeding education included hands-on
assistance and education from maternity ward nurses, written breastfeeding materials
and access to an International Board Certified Lactation Consultant for breastfeeding
problems.
Intervention group (N = 90).

In addition to routine care as for control group, prenatal, perinatal (and postnatal)

Chapman 2004 (Continued)
peer support services, consisting of at least 1 prenatal home visit to review benefits of
breastfeeding, screen for inverted nipples, discuss breastfeeding myths, positioning and
anticipatory guidance. Breastfeeding video viewed if possible. Additional prenatal visits
if necessary
47/89 (53%) reported a prenatal home visit with the mean visit lasting 69.0 (standard
deviation 57.6) minutes. Participants recall of the prenatal visit was: written brochures
provided (38/42); breastfeeding positions reviewed (37/42); breast pumping information
provided (31/42); breastfeeding video viewed (19/42); breastfeeding myths reviewed
(38/42)

Breastfeeding at 1 month and 3 months.
Notes
Risk of bias
Random sequence generation (selection Low risk Participants were randomised to either the
bias) control group or the intervention group
by means of a computer software pro-
gramme. Cases were entered into a data file
weekly, and SPSS randomly selected, ap-
proximately 50%, pg 898
Allocation concealment (selection bias) Unclear risk It is unclear how allocation concealment
was preserved.
Blinding of participants and personnel Unclear risk Participant and personnel blinding was not
(performance bias) described in the text. In the discussion, au-
All outcomes thors say the study was not double-blind
but no details are provided regarding the
extent of blinding that was done
Blinding of outcome assessment (detection High risk Authors of this report say that “interviewers
bias) were unaware of group assignment at the
All outcomes beginning of each interview”.... pg 901, but
failed to give details of how blinding was
done and the extent to which interviewers
were blinded given the above comment
Incomplete outcome data (attrition bias) Low risk Approximately 20% of participants were
All outcomes lost to follow-up in intervention group and
> 20% loss to follow-up in control group.
However, the reasons for dropout was sim-
ilar across both groups. Fig. 1

Selective reporting (reporting bias) Low risk Primary outcomes reported in study pro-
tocol was exclusive breastfeeding rate (time
frame: 3 months postpartum) while sec-
ondary outcomes were breastfeeding rate
(time frame: 3 months postpartum) and
breastfeeding initiation rate (time frame:
for the duration of the hospital stay, av-
erage equals 3 days). however, study re-
port included different primary outcomes
than planned and more secondary out-
comes were reported
Other bias Low risk The study appears to be free of other

sources of bias.
Chapman 2013
Methods Individual RCT, 206 pregnant, overweight/obese, low-income women and randomly
assigned them to receive SBFPC or standard care. Random allocation was done using
computer software “Each week, the study coordinator used SPSS software to randomly
assign 50% of newly recruited participants to the intervention group”
All analyses were by intention-to-treat.
Participants 206 pregnant, overweight/obese, low-income women <= 36 weeks’ gestation. To be

eligible for the trial, women had to be considering breastfeeding and have a prepregnancy
BMI >= 27.0, >= 18 years, <= 36 weeks’ gestation, singleton pregnancy, absence of
medical conditions interfering with breastfeeding, planning to remain in the area for 6
months postpartum, income, 185% of the federal poverty level, and having telephone
access
Interventions SBFPC intervention promoting exclusive breastfeeding among overweight/obese, low-

income women delivered during prenatal visits, postpartum after delivery, and up to 6
months postpartum. Control group received standard care which included Breastfeeding:
Heritage and Pride (BHP)
Outcomes Primary outcomes - breastfeeding initiation and the rates of exclusive and any breast-
feeding at 2 weeks, 1 month, 3 months, and 6 months postpartum. Secondary outcomes
included infant morbidity (diarrhoea, otitis media, emergency department visits, hospi-
talisation), maternal amenorrhoea, and breastfeeding intensity
Notes
Risk of bias

Random sequence generation (selection Low risk Random allocation was done using com-
bias) puter software “Each week, the study co-
ordinator used SPSS software to randomly
assign 50% of newly recruited participants
to the intervention group”, e163
Allocation concealment (selection bias) Unclear risk It is unclear how allocation concealment
was preserved, e163
Blinding of participants and personnel Unclear risk Participant and personal blinding was not
(performance bias) described in the text, e163
All outcomes
Blinding of outcome assessment (detection High risk Data collectors were not completely
bias) blinded. “The interviewer was not in-
All outcomes formed of participants’ group assignment
but was not completely blinded”, e164
Incomplete outcome data (attrition bias) Unclear risk Approximately 20% of participants were
All outcomes lost to follow-up in intervention group and
> 20% loss to follow-up in control group,
e165
Selective reporting (reporting bias) Low risk Primary outcomes reported in study pro-
tocol was exclusive breastfeeding rate (time
frame: 3 months postpartum) while sec-
ondary outcomes were breastfeeding rate
(time frame: 3 months postpartum) and
breastfeeding initiation rate (time frame:
for the duration of the hospital stay, av-
erage equals 3 days). however, study re-
port included different primary outcomes
than planned and more secondary out-
comes were reported
Other bias Unclear risk Baseline characteristics of participants were

different wherein the intervention group
was significantly younger and differed in
delivery mode, compared with controls,
e165

Coombs 1998
Methods Allocation method was an opaque container filled with 100 tags (50 - experimental
group; 50 - control group). Following greater selection of women to the control group, a
statistician calculated the number of control tags to be removed to bias further selection
in favour of intervention tags until groups were balanced
Participants 200 pregnant women, age 18 years or more, literate, no medical conditions likely to make
breastfeeding difficult, willing to consider using the manual and to undertake interview
about breastfeeding
Those who agreed to participate after the interview differed significantly from those who
declined in terms of parity, breastfeeding knowledge, attitudes, confidence, and intention
to breastfeed
Interventions Experimental group (N = 104).

Received the self-help manual 7 weeks before delivery during standard prenatal breast-
feeding counselling from nutritionist. The manual was modelled on successful self-help
smoking cessation interventions to reduce cigarette smoking among low-income preg-
nant women using cognitive behavioural theory. Received a total of 2 prenatal interviews
and 2 postnatal interviews.
Control group (N = 96).

Standard prenatal breastfeeding counselling from nutritionist. No manual. Received a
total of 2 prenatal interviews and 2 postnatal interviews
Outcomes Exclusive breastfeeding at hospital discharge or if breastfeeding initiated later, exclusive

breastfeeding within 1 week
Notes To determine if a self-help manual assisted low-income pregnant women to prepare for,
initiate and maintain breastfeeding
Risk of bias
Random sequence generation (selection Unclear risk “immediately following enrolment, the
bias) women were randomised into either the
treatment of control group”. No further
details regarding how randomisation was
achieved was provided, pg 204
Allocation concealment (selection bias) Unclear risk No descriptions were given regarding allo-
cation concealment.
Blinding of participants and personnel High risk Inadequate due to non-blinding. Partici-
(performance bias) pants were not blinded to treatment group
All outcomes and authors discuss the bias arising from
participants knowledge of study group sta-
tus before intervention (manual distribu-
tion), pg 207

Coombs 1998 (Continued)
Blinding of outcome assessment (detection Unclear risk Not clear if those assessing outcomes were
bias) blind to group allocation. The study out-
All outcomes comes were assessed by maternal self-report
through interviews but authors do not say
if outcome assessors were blinded or not
Incomplete outcome data (attrition bias) High risk Inadequate, 23/104 lost from the inter-
All outcomes vention group and the study involved 200
women who were randomised to either the
treatment or control group (treatment - 96,
control - 104). Overall, there was a 25% at-
trition rate (23 dropout in treatment group
(24%) and 26 (25%) in the control group)
. Reasons for dropout were provided in
the text, but could not be compared across
groups as only aggregate percentages were
provided, pg 204-205, fig 1 26/96 from the
control group (24.5% overall)
Selective reporting (reporting bias) Unclear risk Study protocol was not available to assess
the prespecified outcomes
Other bias Unclear risk The baseline characteristics of study partic-

ipants were not described in sufficient de-
tail to be able to assess if there were differ-
ences between women enrolled in the treat-
ment and control groups
Edwards 2013a
Methods Individual RCT of community doula home visiting. Doulas provided home visits and
support during childbirth. Data were obtained from medical records and maternal in-
terviews at birth and 4 months postpartum. Intent-to-treat analysis used
Participants Low-income, African-American mothers (n = 248) under the age of 22 years. Participants
were recruited when they were less than 34 weeks pregnant and if they were planning to
deliver at the affiliated hospital
Interventions Intervention-group mothers received services from paraprofessional doulas: specialised

home visitors trained as childbirth educators and lactation counsellors. Doulas provided
home visits from pregnancy through 3 months postpartum, and support during child-
birth. Mothers in the community doula intervention group received an average of 10
prenatal and 12 postpartum home visits. A doula was present at the hospital for the birth
for 81.5% of the intervention group infants. Control group mothers received usual care
Outcomes Infant-feeding practices including breastfeeding initiation, breastfeeding duration, tim-

ing of introduction of complementary foods

Edwards 2013a (Continued)
Notes
Risk of bias
Random sequence generation (selection Low risk Randomisation took place in blocks of 4,
bias) 6, or 8, with equal numbers assigned to
the intervention and control groups within
each block. A biostatistician prepared a set
of opaque envelopes, each labelled with a
subject ID number and containing a group
assignment. Comment: prepared by a bio-
statistician, likely random sequence genera-
tion. Information obtained from the ’Ran-
domisation’ section, pg S162
Allocation concealment (selection bias) Low risk “A biostatistician prepared a set of opaque
envelopes, each labelled with a subject
ID number and containing a group as-
signment.” Information obtained from the
’Randomisation’ section, pg S162
Blinding of participants and personnel High risk Participants and researchers were not
(performance bias) blinded to group allocation. Information
All outcomes obtained from the Randomisation section,
pg S162
Blinding of outcome assessment (detection Unclear risk Data on breastfeeding attempts were col-
bias) lected by mother report at the hospital the
All outcomes second morning after the birth and from
review of the nursing notes in the mother’s
medical chart after the mother’s discharge.
Comment: unclear if research staff were
blinded to group assignment. Information
obtained from the Outcome measures sec-
tion, pg S163
Incomplete outcome data (attrition bias) Low risk Only 1 participant lost to follow-up in each
All outcomes group. Information obtained from Figure
1, pg S164
Selective reporting (reporting bias) Unclear risk Trial was registered 2 weeks before the pa-
per was accepted for publication. Retro-
spective registration so a priori outcomes
unclear
Other bias Low risk No other obvious bias.

Edwards 2013b
Methods This is a pilot/feasibility individual RCT to explore the acceptability of a multi-racial,

computer-animated, female, laptop-based Computer Agent designed to improve exclu-
sive breastfeeding rates among mothers interested in breastfeeding. The Computer Agent
was modelled on a live counterpart, an International Board Certified Lactation Consul-
tant. The setting for the intervention was primarily the outpatient offices of Obstetri-
cians/Gynaecologists affiliated with the USA-based hospital
Participants 15 participants (7 in the intervention group, 8 in the control group) completed this study.
Eligible women were primiparas, in their third trimester of pregnancy with a singleton
fetus, 18 years of age or older, English-speaking, had internet access, and were interested
in breastfeeding
Interventions Control arm: the control arm received the standard care relating to breastfeeding. At the
time of this study, that included an optional prenatal breastfeeding class, information on
the benefits of breastfeeding from obstetric offices, encouragement to put the baby to the
breast within the first hour of life, education by all staff on management of breastfeeding,
and lactation consultations once per day or more as needed
Intervention arm: the intervention arm received all aspects of the control arm, plus access
to the Computer Agent to access additional information about breastfeeding. The Com-
puter Agent was used prenatally during a third trimester office visit and perinatally at
hospital discharge. The Computer Agent was designed to present breastfeeding informa-
tion and support focusing on the benefits of breastfeeding and motivational interviewing
techniques prenatally. Dialogue was customised to each participant and the programme
maintained memory of the subject’s demographics (name, baby’s name and sex)
Outcomes Intent to exclusively breastfeed, attitudes toward breastfeeding (as measured by the Iowa
Infant Feeding Attitudes Scale), breastfeeding self-efficacy (as measured by the Breast-
feeding Self-Efficacy Scale Short Form)
Notes This Cochrane Review does not include outcome data from this primary research article
Risk of bias
Random sequence generation (selection Unclear risk The authors state that they used “blocked
bias) randomization, with a blocking factor of 4”
(pg 1964) but do not describe the random
sequence generation
Allocation concealment (selection bias) Unclear risk The authors state that they used “sealed en-
velope[s]” (pg 1964). It is unclear whether
these envelopes were opaque
Blinding of participants and personnel High risk Due to the nature of the intervention, it
(performance bias) would not be possible to blind participants.
All outcomes The authors describe the distribution and
utilisation of the tablet laptops in physi-
cians’ offices prenatally as “somewhat cum-

Edwards 2013b (Continued)
bersome.” This was problematic for office

staff. It is unclear whether these office staff
were part of the research team. If the re-
searchers were managing the distribution
of the tablet laptops, there is potential for
performance bias to be introduced
Blinding of outcome assessment (detection High risk The authors state (pg 1965) that when sub-
bias) jects participating in the study were admit-
All outcomes ted to the hospital, the “study staff visited
them on the birth day of their baby to com-
plete questionnaires and access the Com-
puter Agent (if assigned to that arm).” This
description suggests that study staff collect-
ing data were aware of study-arm alloca-
tion. All enrolled participants were also vis-
ited by study staff at hospital discharge to
collect outcome data
Incomplete outcome data (attrition bias) Low risk Outcome data are available for 13 of the 15
All outcomes participants.
Selective reporting (reporting bias) Unclear risk No protocol or evidence of predefined out-
comes to judge this domain
Other bias Unclear risk No other obvious sources of bias.
Efrat 2015
Methods Individual RCT. Lactation educators (undergraduate students who completed a semester-
long lactation education course and 10 hours of post-course training) developed a rela-
tionship with women prenatally and then phoned mothers regularly postnatally. Data
relating to the factors associated with breastfeeding were collected during the third
trimester. Breastfeeding outcome data were collected at 72 hours, 1 month, 3 months,
and 6 months postpartum. Outcome data were collected by research assistants who used
a phone questionnaire to collect data from the control and intervention group partici-
pants
Participants 289 pregnant, low-income Hispanic women. Women were 26-34 weeks’ gestation at
recruitment, medicaid recipients, self-identified Hispanic, available via telephone, and
not already assigned to a WIC peer counsellor
Interventions Lactation educator-implemented prenatal and postpartum phone-based breastfeeding

education and support. The intervention entailed 4 prenatal and 17 postpartum phone
calls (first call initiated when mothers were in the third trimester of pregnancy and the
last call when mother was 6 months postpartum). The intervention participants were
also provided with the lactation educator’s phone number so they could contact her more
frequently if need be

Efrat 2015 (Continued)
Outcomes Breastfeeding initiation, duration, and exclusivity.
Notes
Risk of bias
Random sequence generation (selection Low risk “… randomised to either the control or in-
bias) tervention group using computer software.
” Comment: the authors do not specify that
the sequence was computer-generated but
it likely was. Information pg 427
Allocation concealment (selection bias) Unclear risk Unclear whether investigators could pre-
dict group allocation
Blinding of participants and personnel High risk Research assistants and mothers were not
(performance bias) blinded to treatment allocation
All outcomes
Blinding of outcome assessment (detection High risk The study protocol prohibited research as-
bias) sistants from providing the control group
All outcomes participants with any breastfeeding educa-
tion or support and also required that they
use the same data collection strategy tech-
niques when collecting data from partici-
pants in both groups. Comment: as previ-
ously mentioned, research assistants were
not blinded to treatment allocation. Infor-
mation pg 427
Incomplete outcome data (attrition bias) High risk Unclear why there are data for breastfeed-
All outcomes ing initiation for 80 control and 77 in-
tervention mothers. 1 reason for dis-en-
rolling people seems to be “discontinua-
tion of breastfeeding” It is unclear whether
the authors have initiation data on these
women. Data are missing with no explana-
tion as to who or why is missing
Selective reporting (reporting bias) Unclear risk No trial protocol available.
Other bias High risk “Despite randomisation, women in the in-

tervention group had a significantly higher
intention to breastfeed.” Information ob-
tained from the ’Results’ section, pg 431

Flax 2014
Methods Cluster-RCT of an integrated microcredit and community health intervention. Baseline

and final survey interviews were conducted by an independent team of trained data
collectors unaware of the clients study arm assignment
Participants 461 pregnant women in 79 microcredit groups.
Interventions The intervention had 3 components. Trained credit officers led monthly breastfeeding
learning sessions during regularly scheduled microcredit meetings for 10 months. Text
and voice messages were sent out weekly to a cell phone provided to small groups of
microcredit clients (5-7women). The small groups prepared songs or dramas about the
messages and presented them at the monthly microcredit meetings. The control arm
continued with the regular microcredit programme
Outcomes Outcome variables were as follows: 1) exclusive breastfeeding to 1, 3, and 6 months; 2)

initiation of breastfeeding within 1-hour of delivery; and 3) use of only colostrum or
breast milk during the first 3 days of life
Notes
Risk of bias
Random sequence generation (selection Low risk Groups were “randomly assigned to inter-
bias) vention and the other to control using a
Bernoulli random variable generated by 1
of the researchers.” Information obtained
from randomisation and eligibility criteria
section, pg 1121
Allocation concealment (selection bias) Unclear risk The authors specify that “monthly meeting
groups with similar numbers of clients and
pregnant women were paired, with 1 group
randomly assigned to intervention and the
other to control...” Comment: the authors
do not specify whether investigators could
know in advance which study arm a meet-
ing group would be assigned to. Informa-
tion obtained from randomisation and eli-
gibility criteria section, pg 1121
Blinding of participants and personnel High risk Participants could not be blinded and
(performance bias) personnel who delivered the intervention
All outcomes could not be blinded due to the nature of
the intervention
Blinding of outcome assessment (detection Low risk “Baseline and final survey interviews were
bias) conducted by an independent team of
All outcomes trained data collectors unaware of the

Flax 2014 (Continued)
clients’ study arm assignment.”

Information obtained from ’Data collec-
tion procedures’ section, pg 1121
Incomplete outcome data (attrition bias) Low risk “At follow-up, 196 (86%) and 194 (84%)
All outcomes clients remained in the intervention and
control arms, respectively.” Comment: data
available for all live births in intervention
group and only missing for 2 live births in
control group (1 maternal death, 1 dyad
moved away)
Information obtained from ’Results’ sec-
tion, 1st paragraph, pg 1122 and Figure 1,
pg 1122
Selective reporting (reporting bias) Unclear risk Comment: could not locate study on “Cur-
rent Controlled Trials” so it’s unclear what
the planned outcomes for this particular
analysis were
Other bias Low risk No other obvious risk of bias.
Forster 2004
Methods A computerised system of biased urn randomisation was accessed by telephone by the
research midwife after written consent was obtained
Analysis undertaken by authors for this review was by intention-to-treat based on data
reported by study authors
Participants Women booking for antenatal care at the Royal Women’s Hosptial in Melbourne, Aus-
tralia, between May 1999 and August 2001. The hospital had been an accredited Baby
Friendly hospital since 1995
Inclusion: booking as public patients, having a first child, 16-24 weeks’ pregnant at
recruitment, able to speak and write in English
Exclusion: physical problems that prevented breastfeeding, chose birth centre or private
obstetric care
Interventions Control group (N = 327).

Received BFHI accredited standard care.
Practical skills intervention (N = 327).
In addition to BFHI accredited standard care, received the offer of a single session of 1.5
hours focusing on practical breastfeeding skills. ’Latching on’ was explained and demon-
strated using dolls and knitted ’breasts’. Breastfeeding complications and management
were discussed. Partners were not present
Attitudes intervention (N = 327).
In addition to BFHI accredited standard care, received the offer of 2 X 1-hour sessions
focusing on changing attitudes to breastfeeding. Partners/significant others were encour-
aged to attend. The first class included information about the advantages of breastfeeding
and explored participants’ views of breastfeeding and their perceptions of the attitudes

Forster 2004 (Continued)
of others. Between classes participants were encouraged to interview their own and their
partner’s mother. The second class included a group discussion based on these interviews,
and discussion of resources for breastfeeding women. Women were encouraged to write
a breastfeeding plan

Breastfeeding and exclusive breastfeeding at 6 months.
Notes Authors concluded that in settings where breastfeeding initiation is high, neither of the
interventions could be recommended as effective
Results not included in the meta-analysis because we considered the control group, BFHI
standard care, had received an intervention that meant we could not compare this control
group with the control groups of other studies in the review
Risk of bias
Random sequence generation (selection Low risk “A computerised system of biased urn ran-
bias) domisation” was used
Allocation concealment (selection bias) Low risk A computerised system of biased urn ran-
domisation was accessed by telephone by
the research midwife to ascertain women’s
group allocation. The research midwife
telephoned the patient and was asked to
follow prompts on the telephone, includ-
ing inputting the woman’s hospital record
number. The random allocation was then
generated
Blinding of participants and personnel High risk It is not stated if women or staff were
(performance bias) blinded but it is stated that women were
All outcomes given a booklet about the study and the in-
tervention was explained to them
Blinding of outcome assessment (detection Unclear risk Data were collected by research midwives
bias) and blinding was not described. It is not
All outcomes clear if the same midwife was responsible
for allocation and data collection
Incomplete outcome data (attrition bias) Low risk 90.3% follow-up.

All outcomes
Selective reporting (reporting bias) Low risk All of the primary outcomes reported in
study protocol were reported in the study.
The secondary outcomes were reported in
a separate paper

Forster 2004 (Continued)
Other bias Unclear risk A smaller percentage of women in the stan-

dard care group received a pension/benefit
as the primary family income (7.2% ver-
sus 16% and 14.6% in the intervention
groups). This difference was not tested for
significance
Hill 1987
Methods Women of different parity were randomised to intervention or control groups

Participants 64 women intending to give birth at the study hospital and keep their infant, and who
gave birth to a healthy infant, and had a telephone or agreed to return the telephone
interview survey by post
95% of the total sample were white women.

Attended a 40-minute lecture including 5-10 minutes for questions and answers; received
a pamphlet with information that reinforced lecture content
Routine breastfeeding classes to all women attending antenatal clinic with no lecture,
discussion, pamphlet or post-test
Outcomes Breastfeeding knowledge scores.

Breastfeeding outcomes: no breastfeeding, any breastfeeding, breastfeeding less than 6
weeks, breastfeeding more than 6 weeks
Notes To determine the effects of a breastfeeding education programme among low-income

pregnant women in Chicago
Risk of bias
Random sequence generation (selection Unclear risk The authors state “Randomization of each
bias) individual in these two subgroups [prim-
ipara/multipara] was carried out” but do
not specify how the random sequence
was generated. Information obtained from
’Method’ section, ’Procedure’ subsection,
pg 149
Allocation concealment (selection bias) Unclear risk No details of allocation concealment avail-
able.

Hill 1987 (Continued)
Blinding of participants and personnel High risk Participants could not be blinded to the in-
(performance bias) tervention. It is unclear whether the inves-
All outcomes tigator delivered the intervention, however,
the investigator was aware of group alloca-
tion as the author states “The investigator
decided to administer the posttest immedi-
ately after the question and answer period
[of the education session] because of avail-
ability of the subjects...” Information ob-
tained from ’Method’ section, ’Procedure’
subsection, pg 149
Blinding of outcome assessment (detection Unclear risk The follow-up interview was conducted by
bias) a “researcher” but it’s not clear if this re-
All outcomes searcher was blinded to group allocation.
Information obtained from ’Method’ sec-
tion, ’Procedure’ subsection, pg 150
Incomplete outcome data (attrition bias) Low risk Breastfeeding initiation data are reported
All outcomes for all participants. Information obtained
from Table 2, pg 151
Selective reporting (reporting bias) Unclear risk No trial registration data available.
Other bias Low risk No other obvious sources of bias.
Hoddinott 2009
Methods RCT with cluster-randomisation. Unit of randomisation and analysis was locality
Participants Pregnant women and breastfeeding mothers registered at GPs in 14 localities (of 66) in
Scotland who gave birth 2002-4. Birth records supplying data n = 9747 in intervention
group and n = 9111 in control group
Interventions Intervention localities were randomised to a policy aim to double the number of local
breastfeeding support groups and to make weekly support groups open to all pregnant
women and breastfeeding mothers. The groups were to be facilitated by health profes-
sionals taking a woman-centred approach and aiming to provide breastfeeding support
and social interaction for women
Control localities received no intervention. Breastfeeding support groups existed in some
control areas
Outcomes Any breastfeeding at birth, 5-7 days and 8-9 months after birth and maternal satisfaction
were secondary outcomes of the study. The primary outcome was number of babies
receiving any breast milk at 6-8 weeks. The study used routinely collected outcome data
for the 2 pre-trial years and the 2 post-trial years
Results were not presented in a way which allowed us to enter them into data and analysis
tables but we have summarised findings in the text

Hoddinott 2009 (Continued)
Notes
Risk of bias
Random sequence generation (selection Low risk Used random number tables.
bias)
Allocation concealment (selection bias) Low risk “An independent statistician used random
number tables to randomise locality pairs
to either intervention or control.” Central
allocation
Blinding of participants and personnel High risk Staff in intervention localities are likely
(performance bias) to have known of the policy intervention
All outcomes and some women in new groups may have
known of it. Other staff and other women
whose outcome data were analysed may not
have known
Blinding of outcome assessment (detection Low risk Researchers analysing primary and sec-
bias) ondary outcomes were reported to be
All outcomes blinded
Incomplete outcome data (attrition bias) Low risk The study authors included all existing rou-
All outcomes tinely collected data in their analyses. Re-
sults were not presented in a way which al-
lowed us to enter them into RevMan data
and analysis tables but we have summarised
findings in the text
Selective reporting (reporting bias) Low risk ISRCTN44857041; All the outcomes re-
ported in the registry were reported in the
text
Other bias Low risk Not known.
Ickovics 2007
Methods Individual-randomised trial. Women at 2 publicly-funded clinics were randomly assigned

to standard individual care or group care
Participants Pregnant women aged 14-25 years attending 2 large obstetric clinics in university-affil-
iated hospitals in the USA. African-American women with limited financial resources
were over represented, which reflected clinic use patterns
Inclusion criteria: less than 24 weeks of gestation, age 25 years or less, no medical
problems requiring individualised care as “high-risk pregnancy” (e.g. diabetes, HIV),

Ickovics 2007 (Continued)
English or Spanish language, and willingness to be randomised. All providers received 2

full days of formal training in Centering Pregnancy group prenatal care
Exclusion criteria: not described.
Interventions Intervention group (n = 394).

Group antenatal care provided by a trained practitioner (e.g. midwife, obstetrician). Ses-
sions first involve self-care assessment of blood pressure and weight and individual prena-
tal assessments by the practitioner. The remainder of the session discussion, education,
and skills building to address explicit learning objectives in prenatal care, child birth
preparation, and postpartum and parenting roles. The full curriculum consists of 10 x
120-minute sessions. All sessions apart from the initial assessment, cervical assessments
in late pregnancy or if health concerns occur, are conducted in this manner
Control group (n = 653).
Individual care. Details not provided.
Outcomes Primary outcomes: gestational age at delivery, birthweight.

Other outcomes included: adequacy of prenatal care, breastfeeding initiation measured
at a 6 month interview, and psychosocial outcomes (pregnancy knowledge, prenatal
distress, readiness for labour and infant care, and satisfaction with prenatal care
Notes Study did not contribute data to the review as the actual number of women who initiated
breastfeeding was not reported. Authors contacted but no response
Risk of bias
Random sequence generation (selection Low risk Computer-generated randomisation se-

bias) quence.
Allocation concealment (selection bias) Low risk “Allocation was concealed from participant
and research staff until eligibility screening
was completed and study condition was as-
signed. A computer-generated randomiza-
tion sequence, password protected to re-
cruitment staff and participants, was used
to assign participants.”
Blinding of participants and personnel High risk “..it was not possible to have treatment
(performance bias) blinded.”
All outcomes
Blinding of outcome assessment (detection Low risk “..all measurement and data collection were
bias) conducted in blinded fashion indepen-
All outcomes dently of the care setting.” The research
team members were independent of prena-
tal care

Incomplete outcome data (attrition bias) High risk The number of women in each group at the
All outcomes postpartum interview was not stated. Only
the total number of women who took part
(n = 783) was reported. This gives a fol-
low-up rate of 74.8% at 6 months. The au-
thors stated there was differential dropout
between group and individual care (P = 0.
95)
Selective reporting (reporting bias) High risk The study protocol does not specify breast-
feeding initiation or satisfaction with pre-
natal care as outcomes
Other bias High risk The original study protocol states this is
a 3-arm trial comparing Centrering Preg-
nancy, Centering Pregnancy Plus and stan-
dard care. The study reported only has 2
arms and it is not reported why there is a
difference or if the 2 intervention arms were
combined
There were significant differences be-
tween intervention and control group with
women in the intervention group hav-
ing significantly greater history of preterm
birth, lower scores for prenatal distress and
a contained a higher percentage of African-
American women
Financial disclosure states that 1 study au-
thor receives approx. USD 3000 per year
from Centering Pregnancy and Parenting
Association Inc and another study author
is the executive director of Centering Preg-
nancy and Parenting Association Inc
Ickovics 2016
Methods Multisite cluster-randomised trial. Clusters were 4 community health centres and 10
hospitals
Participants Pregnant adolescents aged 14-21 years attending an prenatal care visit at 1 of the partic-
ipating clinical sites. The clinical sites were in New York City and predominantly served
low-income women
Inclusion criteria: pregnancy at less than 24 weeks’ gestation, pregnancy not considered
high risk, ability to speak English or Spanish, and willingness to participate in group
prenatal care
Exclusion criteria: not described.


Centering Pregnancy Plus group prenatal care. First visit is an individual clinical assess-
ment and thereafter all care is provided in a group setting. Sessions were facilitated by
a clinician (e.g. obstetrician, midwife) and a co-facilitator (e.g. nurse, medical assistant)
. The 10 X 120-minute sessions first involve self-care assessment of blood pressure and
weight and individual prenatal assessments by the practitioner. The remainder of the
session involves facilitated discussions on many issues related to pregnancy, childbirth,
and postpartum. 4 of the sessions specifically focused on activities to improve sexual self-
efficacy, HIV knowledge, interpersonal sexual communication, perceived risk, and social
norms
Control group (n = 623).
Individual care. Details not provided.
Outcomes Primary outcomes included: gestational age, birthweight and breastfeeding initation.It
is not stated when this was measured and deviates from the protocol which states that
breastfeeding measured at 6 and 12 months is the primary outcome
Secondary outcomes: neonatal intensive care unit admission rates and duration of stay,
incidence of a sexually transmitted infection 12 months postpartum, rapid repeat preg-
nancy and sexual risk behaviours
Notes Does not contribute data to review. Actual numbers of women initiating breastfeeding
not reported (only an as-treated odds ratio presented)
Risk of bias
Random sequence generation (selection Low risk Randomisation was done using a com-
bias) puter-generated sequence in stratified
blocks
Allocation concealment (selection bias) Unclear risk Allocation concealment was not described.
Blinding of participants and personnel High risk The authors recognise that “neither clusters
(performance bias) nor participants could be blinded to study
All outcomes condition”
Blinding of outcome assessment (detection Unclear risk Interviews were conducted by providing
bias) participants with headphones to spoken
All outcomes questions on a display screen and trained
staff reviewed maternal and child medical
records to extract data. It is not stated if
these staff were blinded
Incomplete outcome data (attrition bias) High risk 93.9% of the intervention group and 92.
All outcomes 3% of the control group were followed up,
however, breastfeeding initiation was re-
ported as an as-treated analysis

Selective reporting (reporting bias) High risk Actual study reports breastfeeding initia-
tion but protocol states breastfeeding at 6
and 12 months as outcomes. Clinical trial
registration number: NCT00628771
Other bias Low risk Only significant difference at baseline was

that women in the intervention group were
more likely to be married
ISRCTN47056748
Methods RCT (n = 182 randomised).
Participants Inclusion: primigravid women attending for antenatal care at 20 weeks’ gestation, in-
tending to give birth at the study hospital
Exclusion: women who had started the ’young mums’ parentcraft programme prior to
the 20 weeks’ visit; vulnerable women (e.g. women who did not speak or understand
English); mothers separated from their babies
The setting was a maternity unit in Northern Ireland with Baby-Friendly accreditation
Interventions Intervention (89 randomised)

Women received a ”motivationally enhanced“ version of standard care from staff who
had been trained in a programme called ”Designer Breastfeeding“
Standard care (93 randomised).
At this study hospital, standard care, received by all the study participants, met Baby-
Friendly standards and complied with National Institute for Clinical Excellence (NICE)
guidelines, and consisted of a 2-hour antenatal infant-feeding class, a breastfeeding book
and midwife support for the first 3 weeks after the birth
Outcomes The primary outcome of the study was breastfeeding motivation. Breastfeeding initiation,
exclusive breastfeeding at discharge, and 3 weeks postpartum were secondary outcomes.
Breastfeeding initiation was defined as giving 1 breastfeed or 1 episode of expressed breast
milk
Notes Authors concluded that the study provided preliminary evidence that motivation to
breastfeed can be increased through routine instruction
Risk of bias
Random sequence generation (selection Unclear risk Not described.

bias)
Allocation concealment (selection bias) Unclear risk The authors provided further detail: ”The only way in
which we could conceal group allocation at the recruit-
ment phase and yet develop a process by which delivery
suite midwives would be able to know 3 months later

ISRCTN47056748 (Continued)
to which of two postnatal environments to transfer the

mother and baby“ was as follows: ”A sampling frame was
generated using SPSS 11.5. Numbers 1-250 were en-
tered into the spreadsheet and the following commands
selected: Data - Select cases - Random sample of cases -
Approximate 50% of cases - Unselected cases filtered (1
intervention group, 0 control group). Groups 1 and 0
were then colour coded. The random sampling output
was transferred onto a table with each number replaced
with the appropriate colour of sticker to indicate group
membership - as women gave consent to participate the
next coloured sticker on the sampling frame was placed
on her notes
Blinding of participants and personnel Unclear risk The authors state: “Neither the researcher, nor the re-
(performance bias) search participants could predict their allocated treat-
All outcomes ment”
Blinding of outcome assessment (detection Unclear risk Described as single-blind.

bias)
All outcomes
Incomplete outcome data (attrition bias) Unclear risk (Report pg 18 Fig 3) 234 assessed for eligibility, 182
All outcomes consented and randomised and 144 completed (79%).
Dropouts reported by group but not all with reasons. 57/
93 (61%) randomised to the intervention were known
to have initiated breastfeeding, compared with 53/89
(60%) randomised to the control group
Selective reporting (reporting bias) Unclear risk No trial registration is available.
Other bias Unclear risk Not enough information to judge.
Kellams 2016
Methods Individual randomised trial involving 522 low-income women. A computer-generated

block randomisation sequence using random block sizes, stratified by prenatal clinic, was
used. Sealed, opaque envelopes, which the research assistant opened just prior to loading
the video for the participant to view were used to allocate women to groups
All analyses were conducted on an intention-to-treat basis.
Participants 522 low-income women of 24 to 41 weeks’ gestation who were WIC eligible could
participate in the trial. Women were excluded if they had multiple-gestation pregnancy,
any known contraindication to breastfeeding (e.g. HIV infection, drug use, or receipt of
chemotherapy), or their primary language was not English
Interventions 25-minute educational breastfeeding video (Better Breastfeeding, Injoy Productions,

2008) viewed during the prenatal period in waiting room/examination room while the
participant waited to be seen by the physician or nurse practitioner. Control group

Kellams 2016 (Continued)
received a 20-minute educational video about nutrition during pregnancy (Healthy

Pregnancy Nutrition, Injoy Productions, 2007)
Outcomes Primary outcomes: the initiation of breastfeeding and the exclusivity of breastfeeding
during the newborn hospital stay
Notes
Risk of bias
Random sequence generation (selection Low risk A computer-generated block randomisa-

bias) tion sequence using random block sizes,
stratified by prenatal clinic, was used, pg
154
Allocation concealment (selection bias) Low risk Sealed, opaque envelopes, which the re-
search assistant opened just prior to loading
the video for the participant to view were
used to allocate women to groups, pg 154
Blinding of participants and personnel High risk Personnel were not blinded to the interven-
(performance bias) tion as viewing of the video was done in the
All outcomes examination and/or waiting room, pg 154
Blinding of outcome assessment (detection Low risk Data reported were abstracted from medi-
bias) cal records, and research assistants abstract-
All outcomes ing the data were blinded to the partici-
pant’s group assignment, pg 157
Incomplete outcome data (attrition bias) High risk Only 64% of eligible women were enrolled
All outcomes in the study and reasons for non participa-
tion was not provided, pg 154
Selective reporting (reporting bias) Unclear risk Study protocol was not available.
Other bias Low risk Although there were some differences

in baseline characteristics of participants,
these differences are unlikely to influence
review outcome of interest. Women in the
control group were more likely to live with
a partner or other adult while women in
the intervention group were more likely to
live with a parent, pg 154

Kirkwood 2013
Methods Cluster-randomised controlled trial designed to test the effect of the home-visits strategy
in Ghana delivered by the existing CBSVs. Clusters were made up of districts and towns
Participants All pregnancies to women of reproductive age (15-45 years) that ended in a livebirth or
stillbirth between November 2008 and December 2009, and data for pregnancies, births,
and deaths gathered through the surveillance system established for the ObaapaVitA trial
of vitamin A and maternal mortality and continued for the Newhints trial were used
Interventions Training the CBSVs in the 49 intervention zones to identify pregnant women in their
community and followed by 2 home visits during pregnancy and 3 visits after birth
on days 1, 3, and 7. CBSVs counselled women and their families to promote essential
newborn-care practices, weigh and assess babies for danger signs, and refer sick newborn
babies as necessary
Outcomes Primary outcomes were all-cause NMR and coverage of key essential newborn-care
practices. Secondary outcomes were age-specific and cause-specific NMRs
Notes Other outcomes not clearly stated were included in the report. However, protocol indi-
cates CBSVs training/counselling included training on all such behaviour outcomes re-
ported in the article (clinicaltrials.gov/ct2/show/record/NCT00623337) “The primary
behaviour outcomes were the percent ages of mothers practising the Newhints recom-
mended behaviours. The data were extracted from the birth form administered at the first
surveillance visit after birth; the form included questions about the pregnancy, delivery,
and newborn-care practices promoted by Newhints,” pg 2187
Risk of bias
Random sequence generation (selection Low risk “Computer-generated restricted randomi-

bias) sation was then done in a one-to-one ra-
tio by an independent epidemiologist us-
ing stratified sampling...” pg 2186
Allocation concealment (selection bias) Low risk Allocation was done by an independent
epidemiologist...pg 2186
Blinding of participants and personnel High risk Participants and personnel in the interven-
(performance bias) tion zones were not blinded: “Community-
All outcomes wide meetings were then organised by the
district health management and Newhints
teams during July and August, 2008, and
chaired by the community chiefs. Their
purpose was to introduce the importance
of newborn care to the community; explain
the rationale, content, and structure of the
Newhints intervention; discuss the impor-
tance of community support for its success;
and present the trained CBSVs with their

Kirkwood 2013 (Continued)
Newhints polo shirt, briefcase, and certifi-

cate,” pg 2186
Blinding of outcome assessment (detection Unclear risk “The data were extracted from the birth
bias) form administered at the first surveillance
All outcomes visit after birth...” pg 2187
Incomplete outcome data (attrition bias) Low risk Data on early initiation of breastfeeding
All outcomes are available for > 96% of liveborn infants
in both the intervention and the control
group, Table 2
Selective reporting (reporting bias) Low risk Although early initiation of breastfeeding
wasn’t explicitly stated as a secondary out-
come in the trial protocol, the content of
the 3rd trimester visit of the CBSVs in-
cluded advice to breastfeed the baby im-
mediately after delivery. Thus, it is clear
that this is an outcome the authors were
interested in. Trial protocol (published
2010: www.trialsjournal.com/content/11/
1/58), Table 1
Other bias Low risk No other obvious source of bias.
Lewycka 2013
Methods 2 x 2 factorial cluster-RCT. 48 equal-sized clusters were randomly allocated to 4 groups
Participants 55,931 women in Mchinji district in rural Malawi.
Interventions 1 group received a “women’s group” intervention, 1 group received “peer counsellors”, 1
group received both interventions and the control group received neither. 24 facilitators
guided groups through a community action cycle to tackle maternal and child health
problems. 72 trained volunteer peer counsellors made home visits at 5 time points during
pregnancy and after birth to support breastfeeding and infant care
Outcomes Primary outcomes for the women’s group intervention were maternal, perinatal, neonatal,
and infant mortality rates; and for the peer counselling were infant mortality rates and
exclusive breastfeeding rates
Notes
Risk of bias

Lewycka 2013 (Continued)
Random sequence generation (selection Low risk Researchers “...allocated clusters with a
bias) random number sequence generated in
Stata (version 7.0)”. Information obtained
from ’Methods, randomisation and mask-
ing’ section
Allocation concealment (selection bias) Low risk Allocation was performed by researchers
who were “not involved in the implementa-
tion of the intervention”. Information ob-
tained from ’Methods, randomisation and
masking’ section
Blinding of participants and personnel High risk Masking of allocation was impossible at
(performance bias) participant level.
All outcomes
Blinding of outcome assessment (detection Low risk Data were gathered independently of pro-
bias) gramme implementation. Information ob-
All outcomes tained from ’Methods, randomisation and
masking’ section
Incomplete outcome data (attrition bias) Low risk Volunteer peer counselling group (82.4%)
All outcomes , no intervention group (83%)
Selective reporting (reporting bias) Low risk The authors state that they tested the in-
tervention effect on primary and secondary
outcomes based on “Previously agreed hy-
potheses”. Information obtained from ’Sta-
tistical analysis’ section, pg 1726
Other bias Unclear risk There were baseline differences between

the intervention and control groups post-
randomisation. Also, the authors note
“[b]ecause women knew their intervention
allocation, behavioural answers were open
to best behaviour bias” on pg 1734
Lindenberg 1990
Methods Randomisation using a table of random numbers for the first 3 months. In the 4th month,
a 3rd group were assigned consecutively (due to ethical and organisational limitations)
to a 2nd intervention group. Results from this group have been excluded from this
Cochrane Review due to the lack of randomisation for allocation.
Participants 512 women were randomised and data are reported for 259 women experiencing a nor-
mal, vaginal delivery with no complications and living in poor urban areas of Managua,
Nicaragua
Lindenberg 1990 (Continued)
Interventions Experimental group.

First 3 months of study: 45 minutes of mother-infant contact immediately after birth
with standardised (uniform) breastfeeding promotion followed by complete separation
until discharge. Standardised breastfeeding promotion consisted of a series of specific
breastfeeding promotional messages.
Control group.
First 3 months of study: complete separation throughout hospitalisation with usual (ad
hoc) breastfeeding promotion. Ad hoc breastfeeding promotion consisted of the routine
infant-feeding information a mother might receive which, given the large volume of
deliveries and short hospital stay, was usually very scant to non-existent
Outcomes Any breastfeeding at 1 week.

Exclusive breastfeeding at 1 week.
Any breastfeeding at 4 months.
Exclusive breastfeeding at 4 months.
Notes To examine the effects of early postpartum mother-infant contact, followed by separation
until discharge, on the incidence and continuation of breastfeeding
Risk of bias
Random sequence generation (selection Low risk “The random assignment of study subjects
bias) was accomplished using a table of random
numbers…” Information obtained from
’Materials and methods’ section, ’Design
and interventions’ subsection, pg 182
Allocation concealment (selection bias) Unclear risk It is unclear whether investigators could
have predicted which group a new partici-
pant would have been allocated to
Blinding of participants and personnel Unclear risk Unclear whether blinding of participants
(performance bias) and providers for delivery of intervention
All outcomes and standardised care was adequate
Blinding of outcome assessment (detection Low risk Outcomes assessors were blind to the
bias) “study hypothesis that breastfeeding is a
All outcomes function of early mother-infant contact”.
Information obtained from ’Materials and
methods’ section, ’Design and interven-
tions’ subsection, pg 182

Lindenberg 1990 (Continued)
Incomplete outcome data (attrition bias) High risk 27% of the original sample of 512 were lost
All outcomes to follow-up due to “postpartum mater-
nal or infant complications or failure to lo-
cate homes for follow-up visits”. The break-
down of these reasons for loss to follow-
up is not provided. However, it is stated
that the “group lost to follow-up was sim-
ilar to the remaining group of mothers in
age and marital status, and were similarly
distributed among the three study groups”.
Information obtained from ’Materials and
methods’ section, ’Sample’ subsection, pg
182
Other bias Unclear risk There were no differences in demographic

characteristics between groups. However,
there were significant differences between
the 3 study groups in infant birthweight
and height, episiotomy rates, anaesthesia,
and premature rupture of the membrane.
It is unclear whether these differences may
introduce bias. Information obtained from
’Results’ section, ’Characteristics of the
study population’ subsection, pg 183
MacArthur 2009
Methods RCT with cluster-randomisation. Unit of randomisation: GP antenatal clinic (n = 66)

. Randomisation stratified by size of antenatal clinic and by midwifery team (n = 8)
providing care at the clinic. Unit of analysis was individual women. Planned sample size
(“just under 3000 women”) not achieved (data received from 2511 women giving birth)
Participants All GPe antenatal clinics in 1 Primary Care Trust within a deprived urban area of Birm-
ingham, UK. In this Trust 90% of births (n = 5500 to 6000) were to women from ethnic
minority groups, with > 25% to women born outside the UK. Data from women giving
birth 1 February to 31 July 2007 were included. 70% of these women were in the lowest
10th for deprivation score
Interventions Antenatal peer support intervention clusters.

The Trust recruited 11 peer support workers for breastfeeding, with personal successful
breastfeeding experience of several months and who were, as far as possible, peers of
women in the clinics in ethnicity and language. Peer supporters received 8 weeks training,
based on the UNICEF baby-friendly breastfeeding management course, that addressed
cultural beliefs and barriers appropriate to the local population. The planned level of peer
support was an initial introduction in the antenatal clinic followed by at least 2 contacts,
at 24-28 and around 36 weeks’ gestation, including at least 1 home visit. The purpose

MacArthur 2009 (Continued)
of the contacts was to provide advice and information in the benefits of breastfeeding
and to support women with particular cultural barriers or concerns. The duration of
sessions was to be based on need. All pregnant women registered with GP antenatal
clinics allocated to the intervention received, in addition to usual care, an offer of contact
with a peer support worker
Control clusters.
Standard antenatal care including usual information and advice from midwives on breast-
feeding, without input from community peer support workers
Outcomes Initiation of breastfeeding defined as “a positive response to whether the infant had had
breast milk either at the time of delivery of by the time of hospital discharge, as recorded
in the hospital records”
Notes Type of intervention: antenatal 1:1 peer support contact with individual women
Risk of bias
Random sequence generation (selection Low risk “Randomisation was stratified by size of an-
bias) tenatal clinic and by midwifery team and
undertaken using a computer program.”
Allocation concealment (selection bias) Low risk “Randomisation was undertaken using a
computer program by the trial statistician,
who was blind to the identity of the ante-
natal clinics.”
Blinding of participants and personnel High risk Not possible due to the nature of the inter-
(performance bias) vention.
All outcomes
Blinding of outcome assessment (detection Low risk Data on outcome (and characteristics of
bias) individual women) were obtained anony-
All outcomes mously from the 3 main hospitals where
women attending the study clinics gave
birth
Incomplete outcome data (attrition bias) Low risk Women (4%) who gave birth other than
All outcomes in the 3 main hospitals were not included
in the results. Among women who gave
birth in the 3 hospitals, breastfeeding status
was not known for 57/1140 (5%) women
from clinics randomised to peer support
versus 56/1371 women (4%) from clinics
randomised to standard care

MacArthur 2009 (Continued)
Selective reporting (reporting bias) High risk The trial protocol included 2 secondary
outcomes - breastfeeding continuation rate
at 10-14 days and 6 weeks and breastfeed-
ing at 6 months. These secondary outcomes
were not mentioned in the study report,
neither were they reported on
Other bias Unclear risk There were few differences in the base-
line characteristics of participants in both
groups. the intervention group had more
deliveries in 1 of the 3 hospitals and fewer
African-Caribbean women than the con-
trol group
Muirhead 2006
Methods A 2-group individual RCT. Allocation to control or peer support group was by post-
recruitment concealed allocation, separate for each of 4 strata. Sequences for each stratum
were generated at the start of the trial by computer in blocks of 10. Allocation to control
or peer support group was by post-recruitment concealed allocation
Analysis was by intention-to-treat basis.
Participants 225 women at 28 weeks’ gestation.
Interventions Peer support for breastfeeding. Peer supporters visited participants at least once during
the antenatal period. Peer support was available to women in the intervention group if
they were breastfeeding on returning home from hospital after delivery and if the peer
supporters were informed in time. Control groups received normal breastfeeding support
only
Outcomes Breastfeeding initiation and duration.
Notes
Risk of bias
Random sequence generation (selection Low risk Allocation sequences for each stratum were
bias) generated at the start of the trial by com-
puter in blocks of 10, pg 193
Allocation concealment (selection bias) Unclear risk “Allocation sequences for each stratum
were generated at the start of the trial by
computer in blocks of 10 (that is, five ran-
dom allocations to each of the peer sup-
port and control groups in each different

Muirhead 2006 (Continued)
block of 10) to give approximate numerical

balance between groups. These lists were
never seen by those doing the recruiting.
The allocation of each woman was done
by postrecruitment telephone call to obtain
the next allocation from the lists,” pg 193
Blinding of participants and personnel High risk Authors do not describe blinding but
(performance bias) made the following comment: “There was
All outcomes no post-allocation concealment as once a
woman was allocated to the peer support or
control group this was known to the peer
supporters and others associated with the
trial,” pg 193
Blinding of outcome assessment (detection Unclear risk Outcome assessment was done by mater-
bias) nal self report through questionnaire in-
All outcomes terviews. Although the trial team were not
directly involved in the questionnaire col-
lection, questionnaires were completed in
the presence of health practitioner and that
may have influenced womens’ reporting of
the outcome, pg 194
Incomplete outcome data (attrition bias) Low risk Although reasons for loss to follow-up were
All outcomes not provided, follow-up loss was very low
in both groups (n = 5) fig 1
Selective reporting (reporting bias) Unclear risk Trial protocol was unavailable to assess pre-
specified outcomes
Other bias Unclear risk The baseline characteristics of participants

was comparable and trial appeared to be
free of other sources of bias
Nolan 2009
Methods RCT (pilot study reporting on 50 mother-infant dyads).
Participants Women scheduled for a planned, repeat, caesarean delivery under regional anaesthesia, of
a live singleton at term (at least 37 weeks’ gestation), at a USA hospital with approximately
1500 deliveries per year, a 33% caesarean rate and a 10% repeat, elective caesarean rate
Interventions NIMS intervention.

The intervention took place in the operating theatre and during the immediate postop-
erative period in the obstetric PACU. Protocol components included intra- and postop-
erative environmental manipulation to maintain a maternal-infant spatial distance of not
more than 8 feet, with uninterrupted maternal visual and auditory contact, en face pre-

Nolan 2009 (Continued)
sentation at birth, intraoperative cheek-to-cheek skin contact, a period of uninterrupted

skin-to-skin contact, and mother and infant to be transferred to the PACU together
Control.
Usual care was not standard. Typically, infants were removed from the operating room
promptly after stabilisation and transferred to the PACU in advance of the mother’s
transfer. Most mothers had brief or no physical contact with their infants. Skin-to-skin
contact was not routinely offered in the PACU and initiation of breastfeeding might or
might not occur there
Outcomes Breastfeeding initiation (at birth, by direct observation in the PACU)

Breastfeeding at hospital discharge (from medical records).
Breastfeeding at 4 weeks (by maternal report to a mail survey question ”At 4 weeks of
age, was your baby receiving any feedings with breast milk?)
The study also reported maternal pain and anxiety and infant temperature, respiratory
rate and salivary cortisol levels
Notes Type of intervention: organisation of care - to minimise maternal-infant separation after

repeat elective caesarean birth - not generalisable
Initiation of breastfeeding not defined. Outcome data collected as above
Outcome data.
72 recruited - not reported by group - include without data.
22 excluded (31%) - not reported by group (6 received general anaesthesia, 2 infants
poorly, 14 did not go to the PACU because the PACU was not staffed at the time of the
birth)
50 reported, 25 in each group.
Breastfeeding initiation: NIMS 20/25 versus control 15/25.
Breastfeeding at hospital discharge: NIMS 19/25 versus control 13/25
Breastfeeding at 4 weeks: NIMS 16/25 versus control 8/25.
Risk of bias
Random sequence generation (selection Low risk Randomisation was by coin toss “Mater-
bias) nal-infant dyads recruited were randomly
assigned by the flipping of a coin into con-
trol and experimental treatment groups”
Allocation concealment (selection bias) Unclear risk Group allocation was by a member of the
research team flipping a coin
Blinding of participants and personnel Unclear risk Not possible due to the nature of the inter-
(performance bias) vention.
All outcomes
Blinding of outcome assessment (detection Unclear risk Outcome assessors for breastfeeding initia-
bias) tion and breastfeeding at 4 weeks were not
All outcomes blinded

Nolan 2009 (Continued)
Incomplete outcome data (attrition bias) Unclear risk Analysis was not by intention-to-treat as
All outcomes only those receiving the intervention (that
is, those whose caesarean section opera-
tions were undertaken when the obstetric
postanaesthesia care unit was staffed) were
included in the analysis
Selective reporting (reporting bias) Unclear risk Study protocol was not available.
Other bias Low risk The study appears to be free of other

sources of bias.
Reeder 2014
Methods Participants were randomly allocated to 1 of 3 study arms; no peer counselling, 4 tele-
phone contacts, 8 telephone contacts. Outcomes were reported by mothers to WIC staff
who were not part of the study team. Analysis was by intention-to-treat
Participants 1948 WIC clients recruited during pregnancy who intended to breastfeed or were con-
sidering breastfeeding. There were no exclusions on the basis of age, multiple gestations,
or previous birth history
Interventions Women assigned to the low-frequency peer counselling group were scheduled to receive
4 planned, peer-initiated contacts: the first after initial prenatal assignment, the second 2
weeks before the expected due date, and the third and 4th at 1 and 2 weeks postpartum.
Women in the high-frequency contact group received 4 additional calls at months 1, 2,
3, and 4. The control group received no peer counselling
Outcomes Breastfeeding initiation, duration, and exclusivity.
Notes
Risk of bias
Random sequence generation (selection Low risk “The forms were sorted between Spanish-
bias) and English-speaking clients, after which
they were randomly allocated to 1 of 3
study arms by using a computer-gener-
ated random number function.” Informa-
tion obtained from ’Methods’ section, ’En-
rolment’ subsection
Allocation concealment (selection bias) Unclear risk Information not available in primary article
or supplementary material

Reeder 2014 (Continued)
Blinding of participants and personnel High risk Participants not blinded, peer counsellors
(performance bias) not blinded.
All outcomes
Blinding of outcome assessment (detection Low risk Outcomes were reported by mothers to
bias) WIC staff who were not part of the study
All outcomes team. The study team then collected those
data. In the supplementary material, the
authors clarify that peer counsellors (the in-
vestigators who were unblinded) had no ac-
cess to outcome data. WIC staff collected
breastfeeding outcome data at regular vis-
its, investigators did not have access
Incomplete outcome data (attrition bias) Low risk Less than 10% loss to follow-up per group.
All outcomes Outcome data for breastfeeding initiation
available for 585/635 controls (92.1%),
591/625 intervention group 1 (94.6%),
611/625 intervention group 2 (97.8%)
Selective reporting (reporting bias) High risk Authors stated that their goal was to in-
crease breastfeeding initiation, duration
and exclusivity but did not report breast-
feeding initiation in the paper
Other bias Unclear risk Appears to be more loss to follow-up in the

control group.
Ryser 2004
Methods Random assignment by participants selecting a sealed envelope (not sequentially num-
bered, opacity not specified) to determine assignment to intervention or control group.
Analysis was by intention-to-treat
Participants 54 English speaking pregnant women of 18 years or more, literate, eligible for Medicaid,
access to telephone and stated feeding intention of ’bottle (formula) feed’ or ’undecided’
Marital status and intention to formula feed differed significantly between comparison
groups

Received the Best Start Program (Bryant 1990), presented as a breastfeeding promotion
campaign that aims to allow health professionals to examine women’s misconceptions
and educate them about their specific concerns. It has been marketed since 1992 and its
materials have been used by various programmes, including the SNPWIC Program. In
this study, the researcher used the ’Best Start’ videotapes, training manuals and handouts
to implement the educational programme during 4 prenatal visits (2 more than control
group as visits also included data collection phase).

Ryser 2004 (Continued)

No exposure to Best Start Program. No details of routine breastfeeding promotion ac-
tivities at the physician’s office were provided
Outcomes Any breastfeeding at 1 week postpartum.

Attitudes to breastfeeding.
Social and professional support.
Notes To evaluate the effect of the Best Start Program on breastfeeding attitudes, intention and
initiation in low-income women
Risk of bias
Random sequence generation (selection Unclear risk “Random assignment to groups was ac-
bias) complished by having the subjects select
a sealed envelope to determine their as-
signment to either the experimental group
or the control group.” Comment: unclear
how the random sequence was generated.
Information obtained from ’Methods’ sec-
tion. ’Setting’ subsection, pg 302
Allocation concealment (selection bias) Unclear risk “Random assignment to groups was ac-
complished by having the subjects select
a sealed envelope to determine their as-
signment to either the experimental group
or the control group.” Comment: unclear
if envelope was opaque. Information ob-
tained from “Methods” section. “Setting”
subsection, pg 302
Blinding of participants and personnel High risk Participants were not blinded due to the
(performance bias) nature of the intervention. “All contact
All outcomes with both experimental and control group
subjects was conducted by the researcher
so that standardization of communication
could be optimized.” Thus, the researcher
was not blinded to group allocation. Infor-
mation obtained from ’Methods’ section.
’Intervention’ subsection, pg 302
Blinding of outcome assessment (detection High risk “The researcher telephoned each subject
bias) within 1 week of delivery to ask … ’How
All outcomes did you feed your infant in the hospital?”’
The researcher was not blinded to group
allocation, as per comments above. Infor-
mation obtained from ’Methods’ section.

Ryser 2004 (Continued)
’Setting’ subsection, pg 302
Incomplete outcome data (attrition bias) Low risk Data reported for 27/28 in the control
All outcomes group and 23/26 in the intervention group.
tion, pg 303
Other bias High risk The authors reported that “more exper-
imental group subjects were undecided
about feeding decisions and that more sub-
jects in the control group planned to for-
mula feed.” Information obtained from
’Results’ section, pg 302
Sandy 2009
Methods The Best Beginnings Program is a primary prevention home-visiting programme. This
programme was initially developed as part of the Healthy Families American initiative
Participants Families from 1 of 2 impoverished, predominately Latino census tracts were eligible to
participate in Best Beginnings. Women were eligible to participate if they were pregnant
or had a baby ≤ 3 months. 588 women were recruited to the study. Of these, 281 met
recruitment criteria specific to this analysis: enrolled prenatally, did not drop out prior to
their child’s birth, had a singleton baby, baby was not in the neonatal intensive care unit,
and data were available on infant-feeding practices. Of these 281 mothers, 238 provided
data on infant-feeding method within 1-week of birth
Interventions FSWs provided services to women in both the intervention and control groups through
home visits. Women in the intervention group were visited weekly during pregnancy
and FSWs provided information about prenatal care and infant-feeding methods. If nec-
essary, FSWs also made referrals for mothers in the intervention group to community
agencies for additional support. During the prenatal home visits, mothers in the inter-
vention group received a dedicated breastfeeding promotion intervention that covered
many aspects of breastfeeding. Mothers in the control group were visited by FSWs less
frequently, they were provided with educational material such as booklets and pamphlets
but FSWs did not actively promote breastfeeding among these mothers
Outcomes Rates of any or exclusive breastfeeding among mothers in the intervention group com-
pared with those not exposed to the prenatal intervention. The authors were not explicit
about timing of the outcome measurement in their study aim
Notes This Cochrane Review does not include outcome data from this primary research article
Risk of bias

Sandy 2009 (Continued)
Random sequence generation (selection Unclear risk “[P]articipants were randomly assigned to
bias) either a program group or a control group”.
No further details provided, pg 405
Allocation concealment (selection bias) Unclear risk No descriptions provided in the text.
Blinding of participants and personnel High risk Neither participants not personnel were
(performance bias) blinded. Authors considered this a limita-
All outcomes tion “The lack of double blinding in the
present study is a methodological limita-
tion,” pg 410
Blinding of outcome assessment (detection High risk No descriptions were provided on blind-
bias) ing of outcome assessors. “The possibil-
All outcomes ity of systematic experimenter bias exists
for mother-reported infant-feeding prac-
tices in the present study, since the FSWs
(Family Support Workers) who questioned
mothers about infant-feeding practices
were not blinded to the program versus
control group status of mothers. In ad-
dition, for some mothers, reports about
breastfeeding may have been influenced by
a desire to please their FSWs or give the
“correct” answer,” pg 410
Incomplete outcome data (attrition bias) Unclear risk There are no data on the outcome of inter-
All outcomes est.
Selective reporting (reporting bias) Unclear risk The study protocol was unavailable.
Other bias Unclear risk The characteristics of women in interven-

tion and control groups were not described
Serwint 1996
Methods Random number table with blocks of 10 to assign participants. Allocation of women to a
paediatrician was not completely random as based on paediatrician availability according
to mother’s due date
Analysis undertaken by authors for this review was by intention-to-treat based on data
reported by study authors
Participants 156 nulliparous women, > 18 years, between 8 and 28 weeks’ gestation, who had not yet
selected a paediatrician or wanted their infant to receive paediatric care at the hospital-
based paediatric clinic
Both experimental and control groups comprised 91% of African-American women
Serwint 1996 (Continued)

In addition to routine care, received a scheduled prenatal visit between 32 and 36 weeks’
gestation at a hospital-based clinic with the infant’s future paediatrician. The clinic
was in an urban academic medical centre where mothers received their obstetric care.
Prior to visits, paediatricians received training in counselling parents of newborn infants
and breastfeeding techniques/promotion. During visits, paediatricians recorded data on
timing of pregnancy, preparation for the infant, involvement of father, social support and
maternal medical history. Parents-to-be were counselled on feeding options, advantages
of breastfeeding, infant car safety, circumcision and access to paediatric healthcare.

Similar management except no prenatal paediatric visits.
Outcomes Breastfeeding intention before prenatal visit.

Breastfeeding initiation at birth.
Breastfeeding at 30 days postpartum.
Breastfeeding at 60 days postpartum.
Mothers who changed their mind in favour of breastfeeding after enrolment.
Parent-physician relationship.
Notes To assess the impact of prenatal paediatrician visits on breastfeeding decisions of low-
income mothers
Risk of bias
Random sequence generation (selection Low risk “The study design was a randomized con-
bias) trolled trial using a random number table
with blocks of 10 to assign subjects.”
Allocation concealment (selection bias) Unclear risk It is unclear whether investigators could
have predicted which group a new partici-
pant would have been allocated to
Blinding of participants and personnel Unclear risk Personnel were not blinded. It is unclear
(performance bias) whether participants were blinded to their
All outcomes assigned group
Blinding of outcome assessment (detection Unclear risk “Study outcomes concerning health prac-
bias) tices were obtained by maternal interview
All outcomes at enrollment, at the infant’s 2-month visit,
and by review of the infant’s nursery chart.
” It is unclear whether outcome assessors
were blinded to group allocation. Informa-
tion obtained from ’Outcomes’ section, pg
1070

Serwint 1996 (Continued)
Incomplete outcome data (attrition bias) Low risk Breastfeeding initiation data were available
All outcomes for 74/81 participants in the intervention
group and 70/75 participants in the control
group. Losses were explained adequately,
mostly transfer of obstetrician care. Infor-
mation obtained from Table 1, pg 1071
Other bias Low risk “Dyads in the intervention and control

group did not differ with regard to mater-
nal age, education, type of medical cover-
age, week at which prenatal care was initi-
ated, infant gestational age at birth, race, or
rate of vaginal delivery.” No other obvious
source of bias. Data obtained from Table 2,
pg 1071
Srinivas 2015
Methods RCT comparing peer counselling with usual care, with participants stratified based on
Iowa Infant Feeding Attitude Scale. Iowa Infant Feeding Attitude Scale administered
before birth. Those with a score > 57 were considered to have a positive attitude to-
ward breastfeeding. The Iowa Infant Feeding Attitude Scale score was used to stratify
participants according to positive or negative breastfeeding attitude. Study participants
were then randomised within these strata in blocks of 4 participants in a 1:1 ratio to
intervention (peer counselling) or control (usual care) group. Breastfeeding self-efficacy
short form administered within 5 days after birth
Participants Women ≥ 28 weeks’ gestation, ≥ 18 years old, English-speaking, low-income. Women

with a diagnosis that was an absolute contraindication to breastfeeding (HIV/AIDS,
herpes simplex on the breast, tuberculous lesions of the breast) were excluded
Interventions Low-intensity peer counselling intervention beginning prenatally. The peer counsellor
contacted women between 28 weeks and 1 week prior to delivery, additional contacts
were at the mother’s request. Peer counsellor also contacted mothers within 5 days of
delivery, weekly to 1 month, every 2 weeks to 3 months, and once at 4 months (in person
or by phone)
Outcomes Any and exclusive breastfeeding at 1 and 6 months postpartum. Breastfeeding initiation
was considered any breastfeeding attempts after birth. Exclusive breastfeeding was con-
sidered duration infant was only breastfeeding or receiving human milk since birth
Notes
Risk of bias

Srinivas 2015 (Continued)
Random sequence generation (selection Unclear risk Sequence generation was not described in
bias) the paper.
Allocation concealment (selection bias) Unclear risk Allocation concealment not described in
the paper.
Blinding of participants and personnel High risk Unable to blind participants. Unclear
(performance bias) whether investigators were blinded
All outcomes
Blinding of outcome assessment (detection High risk Outcome data were collected by the study
bias) co-ordinator. Study co-ordinator contacted
All outcomes the control group monthly to assess breast-
feeding status so was unblinded to group
allocation. “The study coordinator admin-
istered the exit interview to both groups ei-
ther after the mother stopped breastfeed-
ing or after 6 months of breastfeeding, to
confirm breastfeeding status as well as per-
ceptions on peer counselling or usual care.
” Thus, it seems like the study co-ordina-
tor collected outcome data and was aware
of group assignment. Information obtained
from ’Recruitment and Study enrolment
procedures’ section
Incomplete outcome data (attrition bias) Low risk 85% follow-up for the outcome of breast-
All outcomes feeding initiation. Although it’s unclear
whether there was equal attrition from
groups, the final sample sizes are similar so
it’s likely that it was relatively evenly split.
tion, 1st paragraph
Selective reporting (reporting bias) Unclear risk Could not locate trial registration.
Other bias Low risk No other obvious sources of bias.
Wambach 2011
Methods RCT with 3 groups.
Participants 390 adolescent mothers aged 15-18, expecting first child in second trimester of pregnancy,
planning to keep the child, can speak and write in English, access to telephone. Multiple
pregnancies, preterm births, infants requiring admission to neonatal intensive care and
participants with birth complications that prohibited or delayed breastfeeding beyond

Wambach 2011 (Continued)
48 hours were excluded. Recruited October 2003-Augst 2006 at 7 prenatal clinics and 4
high schools in the Midwestern USA. Most participants were African-Americans on low
incomes. The groups were similar except that more in the intervention group planned
to return to school

Education and counselling based on TPB and developmental theory, and provided by a
lactation consultant-peer counsellor team from the second trimester of pregnancy to 4
weeks postpartum. 2 prenatal classes, lasting 90 minutes and 2 hours, used a previously
tested breastfeeding education curriculum (Breastfeeding Educated and Supported Teen
Club (BEST), Volpe 2000). Peer counsellor prenatal telephone calls provided ongoing
decision-making support and information
Attention control group (n = 128), to control for non-specific effects of treatment
Same amount of education and counselling, focused on healthy pregnancy behaviours
and birth preparation, not on breastfeeding
Usual care group (n = 134).
Received standard care from their respective clinics, which had varying provider types
and birth settings
Outcomes Breastfeeding initiation defined as initiating breastfeeding in the hospital with intention
to provide at least half the infant’s feedings at the breast or with pumped breast milk,
and measured by self-report in hospital
Breastfeeding duration defined as the total number of days the mother breastfed or
provided breast milk
Exclusive breastfeeding.
Notes
Risk of bias
Random sequence generation (selection Unclear risk “Participants were randomly assigned to one of three study
bias) groups: experimental, attention control, or usual care, using a
list of random codes.” No details are available for how codes
were generated
Allocation concealment (selection bias) Unclear risk “Patients were randomly assigned to one of three study groups
using a list of random codes generated by the study biostatisti-
cian.”
Blinding of participants and personnel High risk Non-blinded. Blinding not possible due to nature of the inter-
(performance bias) vention
All outcomes
Blinding of outcome assessment (detection Unclear risk Non-blinded.

bias)
All outcomes

Wambach 2011 (Continued)
Incomplete outcome data (attrition bias) Unclear risk Reported by group with reasons, in such a way that we could
All outcomes report results by intention-to-treat. Follow-up was: intervention
= 77/122 (63%); attention control = 60/115 (52%); usual care
= 64/119 (52%) i.e. not high
Selective reporting (reporting bias) Unclear risk Study protocol unavailable.
Other bias High risk Baseline characteristics were significantly different in the exper-
imental group compared to the other groups regarding plans to
continue school and TPB variables
BFHI: Baby-Friendly Hospital Initiative

BMI: body mass index
CBSVs: community-based surveillance volunteers
FSWs: Family Support Workers
GP: general practice
NIMS: Nursing Intervention to Minimise Separation
NMR: neonatal mortality rate
PACU: postanaesthesia care unit
RCT: randomised controlled trial
SBFPC: specialised breastfeeding peer counselling
SPSS: Statistical Package for the Social Sciences
TPB: theory of planned behaviour
WIC/SNPWIC: Supplemental Nutrition Program for Women Infants and Children
Characteristics of excluded studies [ordered by study ID]
Study Reason for exclusion
Ahmad 2012 Cross-sectional design; not randomised trial.
Ahmed 2008 Premature infants; intervention after the birth.
Aidam 2005 Not concerned with activity intended to increase breastfeeding initiation rates
Anderson 2005 Not concerned with activity intended to increase breastfeeding initiation rates
Andersson 2013 Trial is a quasi-RCT and does not fit the types of studies for inclusion in this review
Babakazo 2015 Intervention was healthcare staff training to evaluate effect of training on duration of exclusive breast-
feeding
Ball 2006 Not concerned with activity intended to increase breastfeeding initiation rates

(Continued)
Ball 2011 Not concerned with activity intended to increase breastfeeding initiation rates
Begley 2011 Focus of study is models of care. Powered for breastfeeding initiation outcome, but no details of
breastfeeding promotion within the description of the intervention. Participants were allowed to move
between intervention groups as deemed necessary
Bica 2014 Intervention took place after birth.
Bishop 1978 No concurrent controls (3 interventions groups, no routine care group). Thus, is not a RCT
Bonuck 2005 Not concerned with activity intended to increase breastfeeding initiation rates
Bonuck 2013 Intervention concerned with breastfeeding duration.
Bottaro 2009 Not concerned with activity intended to increase breastfeeding initiation rates
Byrne 2000 Not concerned with activity intended to increase breastfeeding initiation rates
Carfoot 2001 [pers comm] Not concerned with activity intended to increase breastfeeding initiation rates
Carfoot 2005 Not concerned with activity intended to increase breastfeeding initiation rates
Cattaneo 2001 Not concerned with activity intended to increase breastfeeding initiation rates
Chapman 1986 Not concerned with activity intended to increase breastfeeding initiation rates
Chapman 2011 Not concerned with activity intended to increase breastfeeding initiation rates
Coutinho 2005 Not concerned with activity intended to increase breastfeeding initiation rates
Di Napoli 2004 Not concerned with activity intended to increase breastfeeding initiation rates
Doherty 2012 Primary outcome was exclusive breastfeeding among women who had already initiated breastfeeding.
Not concerned with activity intended to increase breastfeeding initiation rates
Ekstrom 2012 Not concerned with activity intended to increase breastfeeding initiation rates
Feldman 1987 Not concerned with activity intended to increase breastfeeding initiation rates
Forster 2011 Not concerned with activity intended to increase breastfeeding initiation rates
Froozani 1999 Not concerned with activity intended to increase breastfeeding initiation rates
Garcia-Montrone 1996 Non-RCT.
Garmendia 2015 Study protocol only. Primary aims are unrelated to breastfeeding initiation. Thus, not concerned with
activity intended to increase breastfeeding initiation rates

(Continued)
Girish 2013 Intervention was not support or education (breast crawl) and did not focus on improving or increasing
breastfeeding initiation
Gordon 1999 Not concerned with activity intended to increase breastfeeding initiation rates
Graffy 2001 [pers comm] Not concerned with activity intended to increase breastfeeding initiation rates
Grossman 1988 Contacted authors but unable to acquire sufficient information on method of allocation for this update.
Abstract only available
Gurneesh 2009 Not concerned with activity intended to increase breastfeeding initiation rates
Haider 2000 Not concerned with activity intended to increase breastfeeding initiation rates
Hanafi 2014 Quasi-RCT.
Harvey 1996 Not concerned with activity intended to increase breastfeeding initiation rates
Hegedus 2000 Not a RCT (before-after study).
Henderson 2001 Not concerned with activity intended to increase breastfeeding initiation rates
Hirschhorn 2015 Not a RCT (phase II implementation study).
Hives-Wood 2013 Intervention concerned with breastfeeding duration.
Hopkinson 2009 Not concerned with activity intended to increase breastfeeding initiation rates
Howard 2000 Intervention was not for promoting breastfeeding initiation among women
Ijumba 2015 Study population included women with HIV. Thus, did not target the population of interest
Jahan 2014 Intervention was nutrition education, primary outcomes were gestational weight gain and birthweight.
Thus, not concerned with activity intended to increase breastfeeding initiation rates
Junior 2007 Very low birthweight babies, not healthy term babies. Thus, did not target the population of interest
Kaplowitz 1983 From information provided, we could not tell whether or not randomisation had taken place. We have
written to the authors but have not yet received clarification. Thus, not a RCT
Kastner 2005 Postnatal intervention focussed on measures of the mother-child relationship. Not concerned with
Kistin 1990 Quasi-RCT (women were allocated to the intervention group if they attended clinic on Monday, and
to the control group if they attended on Friday)
Kojuri 2009 Not concerned with activity intended to increase breastfeeding initiation rates

(Continued)
Kools 2005 Not concerned with activity intended to increase breastfeeding initiation rates
Kramer 2001 This study (PROBIT) was primarily concerned with activity intended to increase the duration, but not
the initiation, of breastfeeding
Labarere 2011 Intervention concerned with breastfeeding duration.
Lakin 2015 Intervention took place after birth.
Langer 1996 Not concerned with activity intended to increase breastfeeding initiation rates
Langer 1998 Not concerned with activity intended to increase breastfeeding initiation rates
Lavender 2005 Not concerned with activity intended to increase breastfeeding initiation rates
Loh 1997 Quasi-RCT (intervention was delivered in alternate weeks).
Lucchini 2013 The study is not concerned with activity intended to increase breastfeeding initiation rates. The study
aims to encourage exclusive breastfeeding at 2 months; breastfeeding initiation was part of the inter-
vention
MacVicar 1993 Not concerned with activity intended to increase breastfeeding initiation rates
Mahmood 2011 Not concerned with activity intended to increase breastfeeding initiation rates
Martens 2000 Not a RCT (not randomised).
Martens 2001 From information provided, we could not tell whether or not randomisation had taken place. We have
Martin 2013 Only included mothers who had already initiated breastfeeding
Matilla Mont 1999 Not a RCT (before-after study).
Mattar 2007 Contacted authors but unable to acquire sufficient information on method of allocation for this update
Maycock 2013 Intervention was primarily for fathers and not pregnant women. Thus, did not target the population
of interest
McEnery 1986 Not a RCT (no randomisation at the point of analysis).
McInnes 2000 Not a RCT (not randomised).
McLachlan 2016 Compared caseload and standard midwifery care on mode of childbirth. Thus, not concerned with

(Continued)
McQueen 2011 Postnatal intervention focused on duration and exclusivity of breastfeeding; not concerned with activity
intended to increase breastfeeding initiation rates
Moran 2000 Not concerned with activity intended to increase breastfeeding initiation rates
Moreno-Manzanares 1997 Postnatal intervention. At baseline, all the participants had already initiated breastfeeding. Not con-
cerned with activity intended to increase breastfeeding initiation rates
Morhason-Bello 2009 Not concerned with activity intended to increase breastfeeding initiation rates
Morrow 1999 Not concerned with activity intended to increase breastfeeding initiation rates
Nasehi 2012 Early breastfeeding initiation was the intervention, not the outcome. Study aimed to assess the effect
of early breastfeeding initiation on exclusive breastfeeding duration
NCT00393640 Focus is on milk production later in lactation. Not concerned with activity intended to increase breast-
feeding initiation rates
NCT01916603 The trial was not aimed at promoting breastfeeding initiation
NCT02162498 Participants are all HIV-positive. Thus, did not target the population of interest
Nguyen 2014 Primary purpose was to assess programme impact pathway of Alive & Thrive programme in Vietnam
Nikodem 1998 Not concerned with activity intended to increase breastfeeding initiation rates
Noel-Weiss 2006 Not concerned with activity intended to increase breastfeeding initiation rates
Nor 2012 Report is of a qualitative study conducted within the context of a RCT
Oakley 1990 Not concerned with activity intended to increase breastfeeding initiation rates
Page 1999 Not a RCT (not randomised).
Petrova 2009 Not concerned with activity intended to increase breastfeeding initiation rates
Philipp 2004 Not concerned with activity intended to increase breastfeeding initiation rates
Pisacane 2005 Not concerned with activity intended to increase breastfeeding initiation rates
Pobocik 2000 Quasi-RCT (some school principals would not allow recruitment of control subjects)
Prakhin 2001 Not concerned with activity intended to increase breastfeeding initiation rates
Pugh 2007 Not concerned with activity intended to increase breastfeeding initiation rates

(Continued)
Rea 1999 Not concerned with activity intended to increase breastfeeding initiation rates
Redman 1995 Not concerned with activity intended to increase breastfeeding initiation rates
Reifsnider 1996 Not a RCT (not randomised).
Ross 1983 Not concerned with activity intended to increase breastfeeding initiation rates
Rossiter 1994 From information provided, we could not tell whether or not randomisation had taken place. We have
Schafer 1998 Not a RCT (not randomised).
Schlickau 2005 Not concerned with activity intended to increase breastfeeding initiation rates
Schwartz 2015 Intervention took place after birth.
Schy 1996 Not concerned with activity intended to increase breastfeeding initiation rates
Sciacca 1995 Quasi-RCT (randomisation alternate and not concealed).
Scott 1975 Not concerned with activity intended to increase breastfeeding initiation rates
Sellen 2012 Primary purpose was process evaluation of a RCT.
Shaw 1999 Not a RCT (not randomised).
Sisk 2004 Did not target the population of interest.
Spinelli 2013 Women enrolled in a depression treatment programme.
Susin 2008 Not concerned with activity intended to increase breastfeeding initiation rates
Talukder 2016 Intervention was training for traditional birth attendants.
Toma 2001 Not a RCT (not randomised).
Turan 2001 Not concerned with activity intended to increase breastfeeding initiation rates
Turnbull 1996 Not concerned with activity intended to increase breastfeeding initiation rates
Tylleskar 2011 Not concerned with activity intended to increase breastfeeding initiation rates
Vaidya 2005 Not concerned with activity intended to increase breastfeeding initiation rates
van den Bosch 1990 Not concerned with activity intended to increase breastfeeding initiation rates

(Continued)
Vianna 2011 Participants are premature infants in Special Care Baby Units, not healthy term babies
Volpe 2000 Quasi-RCT (randomisation not concealed, comparison groups not concurrent)
Waldenstrom 1994 Not concerned with activity intended to increase breastfeeding initiation rates
Westphal 1995 This was an evaluation of staff training around the 10 steps of the BFHI and did not specifically focus
on breastfeeding initiation
Wiles 1984 Not concerned with activity intended to increase breastfeeding initiation rates
Winterburn 2003 Contacted authors but unable to acquire sufficient information on method of allocation for this update
Winters 1973 Focus is time to initiation of breastfeeding. Not concerned with activity intended to increase breast-
feeding initiation rates and does not report them
Wolfberg 2004 Intervention was primarily for fathers and not pregnant women
Woolridge 1985 Intervention is timing of initiation of breastfeeding. Outcome is milk transfer. Not concerned with
activity intended to increase breastfeeding initiation rates and does not report them
Yotebieng 2015 Intervention was for healthcare staff training on BFHI Steps 1 through 10
Zimmerman 1999 Not a RCT (not randomised).
BFHI: Baby Friendly Hospital Initiative

PROBIT: Promotion of breastfeeding intervention trial
RCT: randomised controlled trial
Characteristics of studies awaiting assessment [ordered by study ID]
Bakhshi 2015
Methods Randomised clinical trial study.
Participants 80 primigravida women attending Mashahd Omlbanin Hospital randomly allocated in to 2 groups (n = 40 per
group)
Interventions The intervention group and the control group received supportive care and routine care, respectively
Outcomes Onset of lactogenesis II.
Notes Language is in Persian, need translation.

Samieizadeh 2011
Methods Unclear.
Participants 210 primiparous mothers.
Interventions Psychosocial support during labour, delivery and the immediate postpartum period provided by a female companion
of choice
Outcomes Duration of labor, time of delivery, Apgar scores, breastfeeding intent and early breastfeeding initiation 1-hour after
birth
Notes Language is in Persian, need translation.
Characteristics of ongoing studies [ordered by study ID]
ISRCTN23019866
Trial name or title Building Blocks - a trial of home visits for first time mothers
Methods Individually-randomised controlled trial.
Participants Young first time mothers (19 yrs old or under).
Interventions Participants will be randomised to either entry into the FNP arm or to the control arm (universal services),
and will be followed up until 2 years after the birth of the child. The whole trial will last 52 months. Interviews
(either face-to-face or by telephone) for both arms of the trial will be at baseline, 34-36 weeks’ gestation and
6, 12, 18, and 24 months after birth.
If participants are selected to join the group that receives the FNP, they will receive visits from a specially
trained ’Family Nurse’. The Family Nurse would normally go to the participants’ home, but can be elsewhere.
The Family Nurse will visit the participant every week for the first month after they join the study, and then
every other week until the baby is born. The Family Nurse will then visit the participant weekly until the baby
is 6 weeks old and then once every 2 weeks until the child is 20 months old. The last 4 visits are monthly
until the child is 2 years old
Outcomes 1. Changes in prenatal tobacco use (maternal measure), measured at baseline and 34-36 weeks’ gestation
interviews
2. Birthweight (child measure), measured at birth (collected afterwards)
3. Emergency attendances/admissions within 2 years of birth, measured at all time points
4. Proportion of women with a second pregnancy within 2 years of first birth, measured at all time points
5. Intention to breastfeed
6. Prenatal attachment
7. Injuries and ingestions
8. Breast feeding (initiation and duration)
9. Language development
10. Education
11. Employment
12. Income/benefits
13. Home (tenure)
14. Health status

ISRCTN23019866 (Continued)
15. Self-efficacy
16. Social support
17. Paternal involvement
Starting date 24/03/2009.
Contact information Dr Mike Robling

Associate Director South East Wales Trials Unit
Department of Primary Care and Public Health
7th Floor Neuadd Meirionnydd
Cardiff University
Heath Park
Cardiff
CF14 4YS
United Kingdom
Notes ISRCTN23019866
Kimani-Murage 2013
Trial name or title MIYCN Intervention Study.
Methods Women will be recruited into the study and randomised to the intervention or control group. Women in
the intervention group will receive regular, home-based counselling on maternal, infant, and young child
nutrition. Mother-infant dyads will be followed up until the child is 1 year old. Mothers will be regularly
assessed on knowledge, attitudes, and practices regarding maternal, infant, and young child nutrition
Participants 780 pregnant women, and the children subsequently born to them, from 2 slums in Nairobi
Interventions In the intervention arm, CHWs will visit pregnant woman roughly once every month up to 34 weeks’
gestation, after 34 weeks’ visits will occur weekly until delivery. After delivery, CHWs will visit the mother
weekly in the first 1 month. CHWs will counsel women during pregnancy and counselling will continue
until 1 year after delivery. Women will be counselled on maternal nutrition, early initiation of breastfeeding,
breastfeeding positions and attachment, exclusive breastfeeding, frequency and duration of breastfeeding,
human milk expression, and the storage and handling of human milk
Outcomes Primary outcome is exclusive breastfeeding for 6 months. Early breastfeeding initiation is listed as a secondary
outcome in Table 2. Other secondary outcomes include breastfeeding and complementary feeding knowl-
edge and attitudes, the duration of any breastfeeding, complementary feeding practices, nutritional status,
morbidity from diarrhoea, and cost-effectiveness
Starting date September 2012.
Contact information Dr Kimani-Murage: ekimani@aphrc.org
Notes

NCT02084680
Trial name or title Intervention trial to measure the effect of individual prenatal information combined with mobile phones
Methods This is a pragmatic community randomised trial. 8 health centres will be randomised to an intervention arm
and 8 will be randomised to a control arm. VHTs were trained for 5 consecutive days on intervention delivery
Participants All women attending their first antenatal consultation prior to 28 weeks’ gestation were eligible to participate,
regardless of parity. There were no exclusion criteria
Interventions The intervention arm will receive VHTs equipped with mobile phones who will make scheduled home visits
to pregnant women. VHTs will discuss birth preparation, signs of problems during pregnancy, obtaining
items necessary for delivery, and newborn care practices
Outcomes Primary outcomes include hygienic cord care, thermal care, early initiation of breastfeeding (within 1-hour
of birth), and avoidance of pre-lacteal feeds
Starting date June 2013.
Contact information Dr Mangwi Ayiasi: rmangwi@musph.ac.ug
Notes
Williams 2015
Trial name or title WASH Benefits.
Methods WASH Benefits is a 7-armed cluster-randomised trial of water, sanitation, hygiene, and nutrition interventions.
This community-based cluster-randomised controlled trial included an infant and young child feeding (IYCF)
behaviour change component
Participants Women in their 2nd or 3rd trimester of pregnancy.
Interventions Nutrition behaviour change communication on breastfeeding and maternal postpartum nutrition practices
Outcomes Early initiation of breastfeeding (less than or equal to 1-hour after birth)
Starting date
Contact information cpstewart@ucdavis.edu
Notes
CHW: community health workers

FNP: family nurse programme
VHT: village health teams

DATA AND ANALYSES
Comparison 1. Healthcare professional-led breastfeeding education and support versus standard care
No. of No. of
Outcome or subgroup title studies participants Statistical method Effect size
1 Initiation of breastfeeding 5 Risk Ratio (Random, 95% CI) 1.43 [1.07, 1.92]
Comparison 2. Non-healthcare professional-led breastfeeding education and support versus standard care
No. of No. of
1 Initiation of breastfeeding 8 Risk Ratio (Random, 95% CI) 1.22 [1.06, 1.40]
1.1 Low-income or 6 Risk Ratio (Random, 95% CI) 1.21 [1.04, 1.40]
minority-ethnic population
1.2 General population 2 Risk Ratio (Random, 95% CI) 1.30 [0.90, 1.88]
2 Early initiation of breastfeeding 3 Risk Ratio (Random, 95% CI) 1.70 [0.98, 2.95]
Comparison 3. Healthcare professional-led breastfeeding education with non-healthcare professional support

versus standard care
No. of No. of
1 Initiation of breastfeeding 2 895 Risk Ratio (M-H, Random, 95% CI) 1.06 [0.88, 1.27]
Comparison 4. Healthcare professional-led breastfeeding education with peer support versus attention control
No. of No. of
1 Initiation of breastfeeding 1 237 Risk Ratio (M-H, Fixed, 95% CI) 1.21 [0.97, 1.51]

Comparison 5. Breastfeeding education using multimedia versus routine care
No. of No. of
1 Initiation of breastfeeding 2 497 Risk Ratio (M-H, Random, 95% CI) 1.16 [0.63, 2.14]
Comparison 6. Early mother-infant contact versus standard care
No. of No. of
1 Initiation of breastfeeding 2 309 Risk Ratio (M-H, Fixed, 95% CI) 1.08 [0.97, 1.20]
Comparison 7. Community-based breastfeeding groups versus no breastfeeding groups
No. of No. of
1 Breastfeeding rate at birth 1 18603 Mean Difference (Random, 95% CI) -0.01 [-0.05, 0.03]

Analysis 1.1. Comparison 1 Healthcare professional-led breastfeeding education and support versus
standard care, Outcome 1 Initiation of breastfeeding.
Review: Interventions for promoting the initiation of breastfeeding
Comparison: 1 Healthcare professional-led breastfeeding education and support versus standard care
Outcome: 1 Initiation of breastfeeding
Study or subgroup log [Risk Ratio] Risk Ratio Weight Risk Ratio
(SE) IV,Random,95% CI IV,Random,95% CI
Brent 1995 0.6548 (0.2251) 19.2 % 1.92 [ 1.24, 2.99 ]
Hill 1987 0.2989 (0.2382) 18.2 % 1.35 [ 0.85, 2.15 ]
ISRCTN47056748 0.0288 (0.1201) 27.9 % 1.03 [ 0.81, 1.30 ]
Ryser 2004 0.7673 (0.2754) 15.7 % 2.15 [ 1.26, 3.70 ]
Serwint 1996 0.266 (0.2281) 18.9 % 1.30 [ 0.83, 2.04 ]
Total (95% CI) 100.0 % 1.43 [ 1.07, 1.92 ]

Heterogeneity: Tau2 = 0.06; Chi2 = 10.20, df = 4 (P = 0.04); I2 =61%
Test for overall effect: Z = 2.41 (P = 0.016)
Test for subgroup differences: Not applicable
0.01 0.1 1 10 100

Favours standard care Favours prof education

Analysis 2.1. Comparison 2 Non-healthcare professional-led breastfeeding education and support versus
standard care, Outcome 1 Initiation of breastfeeding.
Comparison: 2 Non-healthcare professional-led breastfeeding education and support versus standard care
1 Low-income or minority-ethnic population

Chapman 2004 1.3912 (0.2158) 7.5 % 4.02 [ 2.63, 6.14 ]
Chapman 2013 0.0007 (0.0264) 24.5 % 1.00 [ 0.95, 1.05 ]
Edwards 2013a 0.254 (0.1135) 15.3 % 1.29 [ 1.03, 1.61 ]
Efrat 2015 -0.0005 (0.0181) 24.9 % 1.00 [ 0.96, 1.04 ]
MacArthur 2009 0.1044 (0.1243) 14.1 % 1.11 [ 0.87, 1.42 ]
Srinivas 2015 0.5866 (0.5233) 1.7 % 1.80 [ 0.64, 5.01 ]
Subtotal (95% CI) 88.0 % 1.21 [ 1.04, 1.40 ]

Heterogeneity: Tau2 = 0.02; Chi2 = 47.81, df = 5 (P<0.00001); I2 =90%
2 General population
Muirhead 2006 0.055 (0.2675) 5.4 % 1.06 [ 0.63, 1.78 ]
Reeder 2014 0.4299 (0.2349) 6.6 % 1.54 [ 0.97, 2.44 ]
Subtotal (95% CI) 12.0 % 1.30 [ 0.90, 1.88 ]

Total (95% CI) 100.0 % 1.22 [ 1.06, 1.40 ]
Heterogeneity: Tau2 = 0.02; Chi2 = 50.98, df = 7 (P<0.00001); I2 =86%
Test for subgroup differences: Chi2 = 0.14, df = 1 (P = 0.71), I2 =0.0%
0.2 0.5 1 2 5
Favours standard care Favours non-prof educ

Analysis 2.2. Comparison 2 Non-healthcare professional-led breastfeeding education and support versus
standard care, Outcome 2 Early initiation of breastfeeding.
Comparison: 2 Non-healthcare professional-led breastfeeding education and support versus standard care
Outcome: 2 Early initiation of breastfeeding
Flax 2014 0.9555 (0.2477) 33.5 % 2.60 [ 1.60, 4.22 ]
Kirkwood 2013 0.1989 (0.0669) 44.0 % 1.22 [ 1.07, 1.39 ]
Lewycka 2013 0.5481 (0.4197) 22.6 % 1.73 [ 0.76, 3.94 ]
Total (95% CI) 100.0 % 1.70 [ 0.98, 2.95 ]

0.01 0.1 1 10 100

Favours standard care Favours non-prof educ

Analysis 3.1. Comparison 3 Healthcare professional-led breastfeeding education with non-healthcare
professional support versus standard care, Outcome 1 Initiation of breastfeeding.
Comparison: 3 Healthcare professional-led breastfeeding education with non-healthcare professional support versus standard care
Study or subgroup Experimental Control Risk Ratio Weight Risk Ratio

M- M-
H,Random,95% H,Random,95%
n/N n/N CI CI
Forster 2004 296/327 297/327 65.0 % 1.00 [ 0.95, 1.05 ]
Wambach 2011 77/122 64/119 35.0 % 1.17 [ 0.95, 1.45 ]
Total (95% CI) 449 446 100.0 % 1.06 [ 0.88, 1.27 ]

Total events: 373 (Experimental), 361 (Control)
0.01 0.1 1 10 100

Favours standard care Favours prof educ + peer
Analysis 4.1. Comparison 4 Healthcare professional-led breastfeeding education with peer support versus
attention control, Outcome 1 Initiation of breastfeeding.
Comparison: 4 Healthcare professional-led breastfeeding education with peer support versus attention control

n/N n/N M-H,Fixed,95% CI M-H,Fixed,95% CI
Wambach 2011 77/122 60/115 100.0 % 1.21 [ 0.97, 1.51 ]
Total (95% CI) 122 115 100.0 % 1.21 [ 0.97, 1.51 ]

Heterogeneity: not applicable
0.01 0.1 1 10 100

Favours attention control Favours prof educ + peer

Analysis 5.1. Comparison 5 Breastfeeding education using multimedia versus routine care, Outcome 1
Initiation of breastfeeding.
Comparison: 5 Breastfeeding education using multimedia versus routine care

M- M-
H,Random,95% H,Random,95%
n/N n/N CI CI
Coombs 1998 44/73 34/78 46.8 % 1.38 [ 1.01, 1.89 ]
Kellams 2016 169/174 168/172 53.2 % 0.99 [ 0.96, 1.03 ]
Total (95% CI) 247 250 100.0 % 1.16 [ 0.63, 2.14 ]

0.2 0.5 1 2 5
Favours usual care Favours multimedia

Analysis 6.1. Comparison 6 Early mother-infant contact versus standard care, Outcome 1 Initiation of
breastfeeding.
Comparison: 6 Early mother-infant contact versus standard care

n/N n/N M-H,Fixed,95% CI M-H,Fixed,95% CI
Lindenberg 1990 117/136 101/123 87.6 % 1.05 [ 0.94, 1.17 ]
Nolan 2009 20/25 15/25 12.4 % 1.33 [ 0.92, 1.94 ]
Total (95% CI) 161 148 100.0 % 1.08 [ 0.97, 1.20 ]

Heterogeneity: Chi2 = 1.55, df = 1 (P = 0.21); I2 =36%
0.1 0.2 0.5 1 2 5 10

Favours usual care Favours early contact
Analysis 7.1. Comparison 7 Community-based breastfeeding groups versus no breastfeeding groups,

Outcome 1 Breastfeeding rate at birth.
Comparison: 7 Community-based breastfeeding groups versus no breastfeeding groups
Outcome: 1 Breastfeeding rate at birth
Community-
based BF Mean Mean
Study or subgroup grps No BF grps Mean Difference (SE) Difference Weight Difference
N N IV,Random,95% CI IV,Random,95% CI
Hoddinott 2009 (1) 9635 8968 -0.009 (0.0184) 100.0 % -0.01 [ -0.05, 0.03 ]
Total (95% CI) 9635 8968 100.0 % -0.01 [ -0.05, 0.03 ]

Heterogeneity: not applicable
-0.5 -0.25 0 0.25 0.5

Favours no BF grps Favours community BF grps

(1) Difference in the proportion of women practising any breastfeeding just after birth - adjusted for pre-intervention breastfeeding rates and clustering
APPENDICES
Appendix 1. Note 1
Unless otherwise stated, the sources of international breastfeeding data are the WHO Global Strategy for Infant and Young Child Feeding
(WHO 2003), or the WHO Global Databank on Breast-Feeding (WHO Data Bank 1996). The Databank is not comprehensive at
this time and is dependent on data collected by individual countries using a variety of methods or indicators, or both.
Appendix 2. Note 2
Figures presented are not standardised for mother’s age and age at which she completed full-time education, factors strongly associated
with the incidence of breastfeeding. Standardised figures were not available for individual countries. Available data for changes in
breastfeeding rates for England and Wales between 2000 and 2005, when standardised for mother’s age and education, report a real
increase in breastfeeding rates which was not simply due to changes in the sample composition (Bolling 2007).
WHAT’S NEW
Last assessed as up-to-date: 29 February 2016.
Date Event Description
29 February 2016 New citation required but conclusions have not Conclusions not changed.
changed
29 February 2016 New search has been performed Search updated. Methods updated.
In this update, we excluded two trials (Howard 2000;
Lucchini 2013), one of these was included in the pre-
vious review (Howard 2000). We added 22 new tri-
als in this update (Caulfield 1998; Chapman 2004;
Chapman 2013; Edwards 2013a; Edwards 2013b;
Efrat 2015; Flax 2014; Forster 2004; Hoddinott
2009; Ickovics 2007; Ickovics 2016; Kellams 2016;
Kirkwood 2013; Lewycka 2013; MacArthur 2009;
Muirhead 2006; Nolan 2009; Reeder 2014; Sandy
2009; ISRCTN47056748; Srinivas 2015; Wambach
2011).
This update (2016) now includes 28 trials and excludes
125 trials

HISTORY
Protocol first published: Issue 3, 1999
Review first published: Issue 2, 2005
Date Event Description
15 January 2008 Amended Converted to new review format.
4 December 2007 New search has been performed Search updated and 25 new trials identified. We included five new trials,
(Caulfield 1998a; Chapman 2004; Forster 2004a; Ryser 2004; Wolfberg
2004a) and excluded 21 new trials (Aidam 2005; Anderson 2005; Ball 2006;
Bonuck 2005; Carfoot 2005; NCT00393640; Coutinho 2005; Di Napoli
2004; Garcia-Montrone 1996; Grossman 1988; Kools 2005; Lavender 2005;
Mattar 2007; Muirhead 2006a; Noel-Weiss 2006; Philipp 2004; Pisacane
2005; Schlickau 2005; Sisk 2004; Vaidya 2005; Winterburn 2003).
30 May 2006 Amended Corrected data error in Graph 01.01 for Brent 1995. No change to conclu-
sions.
Search updated. Twenty-four new trial reports added to ’Awaiting assessment’
for next update, which is currently being prepared.
CONTRIBUTIONS OF AUTHORS
The 2016 update (which involves new authors):
• Olukunmi Balogun: independent screening, data extraction, quality appraisal, analysis and synthesis of findings, edited and
updated results, and revised the manuscript.
• Elizabeth J O’Sullivan: data extraction, quality appraisal, analysis and synthesis of findings, updated results, and revised the
manuscript.
• Alison McFadden: edited results and discussion, and revised the manuscript.
• Erika Ota: data extraction, quality appraisal, analysis and synthesis of findings, and ’Summary of findings’ tables.
• Anna Gavine: edited results and discussion, and revised the manuscript.
• Christine Dieterich Garner: independent prescreening, data extraction, and quality appraisal.
• Mary Renfrew (contact author): revised the manuscript.
• Steve MacGillivray: applied the study selection criteria, edited results and discussion, and revised the manuscript.
All the authors read and approved the final version to be published.
DECLARATIONS OF INTEREST
Olukunmi O Balogun: none known.
Elizabeth J O’Sullivan: none known.
Alison McFadden: none known.
Erika Ota: none known.
Anna Gavine: none known.
Christine D Garner: none known.
Mary J Renfrew: none known.
Stephen MacGillivray: none known.
SOURCES OF SUPPORT
Internal sources
• Mother and Infant Research Unit, University of Leeds, UK.
External sources
• Canadian Cochrane Child Health Field Bursary Award, Canada.
• York Centre for Reviews and Dissemination, UK.
• Evidence and Programme Guidance Unit, Department of Nutrition for Health and Development, World Health Organization,
Switzerland.
DIFFERENCES BETWEEN PROTOCOL AND REVIEW

Methods updated to current standard text of Cochrane Pregnancy and Childbirth.
We have edited the main outcomes from ’Initiation and duration of any and exclusive breastfeeding’ to:
1. initiation of breastfeeding;
2. early initiation of breastfeeding (within one hour after birth).
We have assessed the quality of the body of evidence using the GRADE approach.
We have redefined our planned subgroup analysis to be based on low-income (or minority-ethnic) population versus the general
population.
In protocol but not review - Types of participants: In order to examine intermediate/process outcomes, other participants exposed to
such interventions, for example partners, health professionals and employers will be considered.
In protocol but not review - Types of outcomes: Process outcomes (health literacy, public policy, social influence).

INDEX TERMS
Medical Subject Headings (MeSH)

∗ Health Education; Breast Feeding [∗ psychology]; Randomized Controlled Trials as Topic
MeSH check words

Female; Humans


Balogun1996Interventions For Promoting The Initiation of Breastfeeding

Caricato da

Informazioni sul documento

Titolo originale

Copyright

Formati disponibili

Condividi questo documento

Condividi o incorpora il documento

Opzioni di condivisione

Hai trovato utile questo documento?

Questo contenuto è inappropriato?

Copyright:

Formati disponibili

Balogun1996Interventions For Promoting The Initiation of Breastfeeding

Caricato da

Copyright:

Formati disponibili

Cochrane Database of Systematic Reviews

Interventions for promoting the initiation of breastfeeding

Interventions for promoting the initiation of breastfeeding (Review)

Interventions for promoting the initiation of breastfeeding (Review) i

Interventions for promoting the initiation of breastfeeding

Cornell University, Ithaca, NY, USA

Editorial group: Cochrane Pregnancy and Childbirth Group.

PLAIN LANGUAGE SUMMARY

Interventions for promoting the initiation of breastfeeding (Review) 3

Population: wom en exposed to interventions intended to prom ote breastf eeding

Risk with standard care Risk with health-

GRADE Working Group grades of evidence

Interventions for promoting the initiation of breastfeeding (Review) 5

Interventions for promoting the initiation of breastfeeding (Review) 6

There were no secondary outcomes included in this review.

Data collection and analysis

Interventions for promoting the initiation of breastfeeding (Review) 7

Interventions for promoting the initiation of breastfeeding (Review) 8

Assessment of the quality of the evidence using the

Interventions for promoting the initiation of breastfeeding (Review) 9

When we identified substantial heterogeneity, we investigated it

Interventions for promoting the initiation of breastfeeding (Review) 10

Interventions for promoting the initiation of breastfeeding (Review) 11

Interventions for promoting the initiation of breastfeeding (Review) 12

Interventions for promoting the initiation of breastfeeding (Review) 13

Interventions for promoting the initiation of breastfeeding (Review) 14

Early initiation of breastfeeding: Three studies evaluated the ef-

Interventions for promoting the initiation of breastfeeding (Review) 16

Population: wom en exposed to interventions intended to prom ote breastf eeding

Risk with standard care Risk with non- health-

Early initiation of Study population average RR 1.70 76,373 ⊕⊕ It is not possible to

GRADE Working Group grades of evidence

Interventions for promoting the initiation of breastfeeding (Review) 19

Interventions for promoting the initiation of breastfeeding (Review) 20

Early mother-infant contact for women with vaginal or caesarean

Further research to evaluate interventions that combine health

Interventions for promoting the initiation of breastfeeding (Review) 21

Interventions for promoting the initiation of breastfeeding (Review) 35

Characteristics of included studies [ordered by study ID]

Methods Randomisation by permuted block.

Interventions Experimental group: (N = 51).

Outcomes Incidence of breastfeeding in hospital.

Notes To determine if a comprehensive breastfeeding promotion programme increased the

Bias Authors’ judgement Support for judgement

Interventions for promoting the initiation of breastfeeding (Review) 36

Other bias High risk Mothers in the intervention group were

Interventions for promoting the initiation of breastfeeding (Review) 37

Interventions 2 x 2 factorial design.

Outcomes Breastfeeding initiation.

Bias Authors’ judgement Support for judgement

Interventions for promoting the initiation of breastfeeding (Review) 38

Other bias High risk Baseline differences in parity, education,

Interventions Control group (N = 75).

Intervention group (N = 90).

Interventions for promoting the initiation of breastfeeding (Review) 39

Outcomes Breastfeeding initiation.

Bias Authors’ judgement Support for judgement

Interventions for promoting the initiation of breastfeeding (Review) 40

Other bias Low risk The study appears to be free of other

Participants 206 pregnant, overweight/obese, low-income women <= 36 weeks’ gestation. To be

Interventions SBFPC intervention promoting exclusive breastfeeding among overweight/obese, low-

Bias Authors’ judgement Support for judgement

Interventions for promoting the initiation of breastfeeding (Review) 41